One of the most common themes the Reds’ broadcasters discuss on a daily basis is the team’s “chronic inability” to hit with runners in scoring position. From listening to the games you probably know by now that the Reds are dead last in the major leagues in AVG w/ RISP. The voices of the Reds on both TV and radio have been harping about batting average with runners in scoring position for several years now. It is a common refrain and a frequent topic of conversation and has been for a long time. Let’s nail some facts to the wall and find out just how concerned we should be about this problem…

AVG RISP 1

We can see in the chart that the Reds are in fact the worst team in baseball in terms of AVG w/RISP this year with a .194 average in that situation. The team the Reds are currently playing, the Kansas City Royals, have been amazing with RISP, batting well over .300 in those critical situations to lead the league by a comfortable margin. Obviously the Royals have been much more efficient at converting their prime scoring opportunities into runs. One would think that excelling at hitting with runners in scoring position would automatically lead to scoring a lot of runs, but that does not appear to be the case. One thing you can observe in the chart above is that there is not much correlation between a team’s AVG w/RISP and their Runs per Game. For example, the Rockies have been 2nd-best with a .302 AVG w/RISP yet they are only 21st in runs per game. Similarly, the Diamondbacks have hit only .245 w/RISP (21st place) yet they are 5th in the majors in runs scored per game. This is true all up and down the chart. 11 of the 30 teams have ranks in these two stats that differ by 10 or more slots. The Reds are 30th in AVG w/RISP yet they are much better at actually producing runs (20th).

Why is that? There are several reasons. The first is that some teams are better at getting runners into scoring position in the first place. The Tigers have generated 437 plate appearances with runners in scoring position, the Angels have generated only 304 plate appearances with runners in scoring position. That means the Tigers have had 44% more opportunities to bat with RISP. The Cubs have produced the 2nd most RISP opportunites with 421, so it is no surprise that they rank 14th in scoring even though their lowly .223 AVG w/RISP ranks 26th. You can succeed with a poor AVG w/RISP if you create more RISP opportunities. The best way to create extra RISP opportunities is to have a high team On-base percentage (OBP).

More reasons why AVG w/RISP does not correlate well with scoring are because you don’t need to have a runner in scoring position in order to score a run. Runners can be knocked in from first base, or they can drive themselves in by hitting a home run. Power-hitting teams that hit a lot of extra-base hits can produce more runs than a singles-hitting team with a higher AVG w/RISP. Teams that hit a lot of extra-base hits will have a high team slugging percentage (SLG). It is also possible to get a hit with a runner in scoring position without actually scoring a run. If a runner is on second base and the batter gets a base hit, the runner might not score. He could stop at third base or he could get thrown out at the plate.

This brings up the question of why are we looking at a form of batting average in the first place? AVG w/RISP is merely a subset of AVG. Longtime readers of Redleg Nation know that batting average is a severely flawed stat that is considered misleading and obsolete by modern baseball standards. Batting average has two major flaws: it ignores walks completely, and it treats all hits the same. Each type of hit, including the walk, has a different value. It is obvious that home runs are much more valuable than singles. Walks are not as valuable as hits but they do often lead to runs. So why use a statistic that totally fails to consider those factors? Since we know that batting average is a ridiculous stat, it therefore follows that batting average with runners in scoring position is also a ridiculous stat. To evaluate a hitter’s performance there are much better statistics than batting average. OBP and SLG are both much more accurate than AVG and both of them correlate with run scoring much better than AVG does. Add them together to get an even more accurate stat called OPS (OBP plus SLG). There are some even better statistics out there on stat-geek sites like FanGraphs and Baseball Prospectus. wOBA and wRC+ are the best available right now but I am sure new ones will be developed.

Here are the Reds’ batting average numbers over the last several seasons:

AVG RISP 2

This chart shows that the Reds almost always finish the season with an AVG w/RISP that is very close to or slightly higher than their overall batting average for the season. This is true around the league going back for many decades. The Reds are scoring more runs per game this year than last year despite hitting 51 points lower with runners in scoring position! Their overall batting average is the same (.238) but despite tanking in AVG w/RISP they are scoring more runs. Why? Because their OPS has risen dramatically. Their OBP and SLG are both way up. They are getting more runners on base and they are getting more extra-base hits. Those factors are much, much more important than their AVG w/RISP.

The league average AVG w/RISP over the 2009-15 period is .257, so the Reds were slightly below average in that stat, but not to the extent that the broadcasters have made it seem. The constant moaning about the Reds’ “chronic inability to hit with runners in scoring position” just is not backed up by the numbers. The Reds have hit just about like we would expect w/RISP — slightly better than their overall batting average. The broadcasters  have perceived the problem to be much worse than it has been. The Reds’ problem scoring runs the last several years has been due to overall mediocrity in all facets of the offensive game, not due to a glaring inefficiency with runners in scoring position.

Historically, AVG w/RISP stats have been slightly higher than the overall batting average. For example from 2007 until today (an 8+ season period consisting of 1.4 million plate appearances) the overall league batting average has been 0.258 while the league batting average with runners in scoring position has been 0.263 over the same period. Why is AVG higher with runners in scoring position than without? Is it because batters bear down and focus more in those situations? No. Is it because batters cut down their swings to make contact so they are more likely to drive the run home? No. The reason is because situations with runners in scoring position are more likely to come with a bad pitcher on the mound than with a good pitcher on the mound. Think about it, which pitchers allow more baserunners? Good pitchers do not allow you to get a runner into scoring position very often, bad pitchers do. A team is more likely to get runners into scoring position vs Jason Marquis than against Johnny Cueto. The proportion of RISP situations vs bad pitchers outweigh the proportion of RISP situations vs good pitchers, so it makes sense that leaguewide RISP batting averages would be higher than the overall batting average numbers.

One more reason to disregard AVG w/RISP numbers: Small Sample Size. Since 2007 leaguewide, only 26% of plate appearances have come with runners in scoring position. So by focusing on RISP stats you are severely reducing the sample size you are evaluating. Chopping stats into small sample sizes leads to wonky numbers that can mislead us into erroneous conclusions. For example, we all know that the day of the week does not affect a batter’s swing. We would not expect the Reds to play any differently on a Tuesday vs a Wednesday vs a Saturday.

AVG RISP 3

Should we use the data in the chart to conclude that the Reds are good hitters on Saturdays and are likely to continue hitting .260 on Saturdays for the rest of the season? Should we believe that the Reds have a “chronic inability” to hit on Fridays? No, of course not. The data here doesn’t mean anything because we are talking about small sample sizes that represent a small percentage of the Reds total at-bats. By focusing on a small sample size, such as the 15% of their games they have played on Saturdays, we become vulnerable to being deceived. We have to decide whether or not the data show a true skill (ability to hit on Saturday) or if it is just a small sample size mirage. When it comes to days of the week we intuitively know that we are looking at a small sample size anomaly here. But AVG w/RISP works the exact same way. People (and Reds’ broadcasters) fail to recognize that AVG w/RISP differs from the team’s overall batting average for the same exact reason that the team’s Saturday batting average differs from their overall batting average: small sample size. Batting ability w/RISP does not represent a distinct skill that is different from overall batting ability. History has shown with startling clarity that major league hitters hit the same whether there are runners in scoring position or not. If you want to predict how the Reds will hit w/RISP the rest of the season you should predict they will hit the same as their overall batting average plus about 5 points. The Reds’ current overall batting average is .238, so we should expect the Reds to hit about .243 w/RISP from now until the end of the season. That prediction will not be spot on, but it will be a lot closer than their current .194 AVG w/RISP.

Conclusion

To answer the question I posed in the title, does it matter that the Reds have been awful w/RISP? The answer is no, not really. The Reds have hit very poorly with runners in scoring position this season. That is a fact. It does NOT mean they have a “chronic inability” to hit with RISP. In the future they are likely to hit w/RISP somewhere near their overall batting average, which is currently .238 for the season. The deviation of their AVG w/RISP from their overall batting AVG is due to small sample size random fluctuation. Hitting w/RISP does NOT represent a separate skill set from hitting without RISP. The Reds have been much better at scoring runs than they have been at hitting with runners in scoring position. They have been able to create a reasonable number of scoring opportunities and they have hit a goodly number of extra-base hits. They still need to score more runs if they want to earn a playoff berth. They have improved their OBP and SLG significantly over last year, and that is why they are scoring more runs than last year despite nosediving w/RISP. Those improvements are nice but not good enough to be a legitimate contender. Their OPS ranks 16th in the majors, which is middle of the pack or league average. League average doesn’t win championships.

Usage of batting average as a tool to evaluate hitters is obsolete and marks the user as behind the times. Citing AVG w/RISP as an important metric is an even more egregious misuse of misleading statistics to draw conclusions that are unsupported by the data. There are better tools available that will not lead to dubious claims and a belief in things that have been disproven.

All statistical data provided by ESPN Trumedia and FanGraphs.

106 Responses

  1. Kurt Frost

    But clutch… And grit… And Marty tells it like it is!?!?

    Great article.

    • greenmtred

      And yet…we all bemoan their inability to score runs and win games. Of course BA and BARISP are flawed stats. But all or most stats are flawed if you want one that is all-inclusive. You can dress the pig up, if you like, and show that given a large enough sample size it evens out (sorry for the mixed metaphor, but I like the imagery), but the Reds have certainly lost games this year that they could have won had somebody gotten a hit with runners in scoring position. I care about wins and losses, even though I’m certain that somebody is going to prove that they are meaningless, dependent stats.

      • MrRed

        Some interesting thoughts. But when it comes to wins and losses with this team, the biggest issue is the lousy pitching both starting pitching and BP. There’s plenty of anecdotal and hard data to support that theory.

      • Steve Mancuso

        We should bemoan their ability to score runs. We just need to blame the right cause. It’s no secret. Walt Jocketty pinpointed it in the offseason. He just didn’t do anything about it. In fact, he made the situation worse. The Reds need to get more runners on base. The ability to get on base is an identifiable and repeatable skill. It is highly correlated with a team scoring runs.

        Batting average isn’t a flawed stat. It measures the ability for the batter to get a hit, given the random variance of where balls land/skill of the defense. That’s an identifiable and repeatable skill. Pete Rose > Willy Taveras. It’s part of evaluating hitters. But too many people cite AVG and then just stop. As though that’s all you need to consider when evaluating a hitter. But it does have meaning.

        Batting average with RISP *describes* the past, but offers no insight into why a team fails. Quoting a team’s batting average on Fridays *describes* previous Fridays. But we understand there is nothing that makes Friday different from other days of the week. So that stat doesn’t offer insight into why a team fails. Imagine a GM who announced that his team wasn’t hitting the ball well on Fridays, so he was going out to find a “Friday AVG guy” (a former Cardinal, to be sure). We’d think the GM nuts.

        Putting the finger on RISP offers a simple but wrong-headed way to figure out what’s the matter with the Reds.

      • WVRedlegs

        Excellent. Thanks for some clarification. I get riled when the SABR guys keep saying BA and RBI’s are meaningless. They aren’t. They do have meaning. But it is also important as you note, to not stop there. Peel back a few more layers with the analytical stats is a better way to evaluate. The deeper any evaluation is, the better it is.

      • wkuchad

        “Longtime readers of Redleg Nation know that batting average is a severely flawed stat that is considered misleading and obsolete by modern baseball standards. … Since we know that batting average is a ridiculous stat, it therefore follows that batting average with runners in scoring position is also a ridiculous stat.”

        I agree WVREDGEGS. You’re not going to win over a lot of people calling their way of thinking ridiculous. There’s no point in it. I agree with most everything in the article. I just wish “SABR guys” could give the facts and analysis without putting down others. I don’t know, maybe it’s just the way I read it.

      • greenmtred

        I largely agree, Steve, but all statistics describe the past (I differentiate statistics from projections). By calling BA flawed, I was intending to acknowledge that it tells far from the whole story. The fact that a batter strikes out with a runner on third is not predictive because it doesn’t mean that he will repeat that sorry outcome the next time he has the opportunity. But if a hitter often hits better with the game on the line than he does when it isn’t, it may indicate something. Clutchiness is widely disparaged here, but not by me. I’ve seen examples in my working life, and don’t believe it to be a non-factor in baseball, even though it is hard to quantify.

      • jdx19

        GREENMTRED. While it is true that statitstics derived from past data technically all describe the past, some stats can describe AND predict. The way to tell if a stat can do that is by getting heaps of season-to-season data for a particular stat and seeing how well it correlated with itself between years.

        As an example, let’s take ERA. These numbers are off the top of my head and illustrative in nature, but they hopefully will get the point across.

        Y2Y Correlation of RA/9: .25
        Y2Y Correlation of ERA: .30
        Y2Y Correlation of FIP: .60

        Based on what we know about correlation, we can factually conclude that FIP is a better number to judge the future on than is RA/9.

        That doesn’t mean FIP doesn’t describe the past, because it does. It measures how often you walked and struck out guys, and how often you gave up home runs. All that happened in the past. But, still, the measure shows a better glimpse into the future.

        If you do the same thing with something like BABIP (again, numbers made up, but relationships accurate)

        Y2Y Correlaton of BABIP: .25
        Y2Y Correlation of AVG: .30
        Y2Y Correlation of OBP: .50

        So it’s not that OBP doesn’t describe the past, beause it does, but it’s simultaneously better at describing the future while also describing the past.

        If anyone is interested in this kind of stuff regarding pitching in A LOT more depth, read this article: http://www.hardballtimes.com/fip-in-context/

      • Steve Schoenbaechler

        Couldn’t have said it better, Steve. All stats show past results. Most all predictions would be made on past results. Thus, predictions are pretty much just a measure of past results.

        As well as, all stats never tell why. Even if you provide another stat that “may” tell why, you may not necessarily be able to tell why that stat justifies the first stat. To tell the “why”, you have to get more into the strategy and/or play of the game, aka take it off the stat table and put it on the playing field.

      • Steve Schoenbaechler

        JDX, nothing and no one can ever predict something will happen. If that was the case, then there would be a lot more of it happening. Now, it can show what it “most likely” to occur. But, that is entirely another matter. There is a world of difference between what is most likely to occur and predicting something to happen.

        Only one thing everyone can predict, everyone that is born will die. Anyone tells you differently is blowing smoke.

  2. big5ed

    Of course it matters. They got hits in 19% of their chances. If they had gotten hits in 28% of those chances, then they would have scored more runs and won more games. They wouldn’t be 8 games behind; they’d be about 5 games behind. If a basketball player hits 38% of his 3s, rather than 29 %, then his team will have a better chance of winning. This is the old ATT ad campaign: More is better than less.

    Batting Average may be crude, and it may or may not be predictive, but it does tell you what actually happened.

    I agree that the a player cannot really be better in a “clutch” setter. But I do think he can be worse when the pressure is on. Remember the look on Calvin Schiraldi’s face on the Shea Stadium mound in the 1986 World Series? I think a good “clutch” hitter is one who generally gives the same quality at-bat in the “clutch” (regardless of the outcome) as he does in an ordinary situation. While this is subjective, the quality of the Reds at bats in big situations right now appears poor, and the slump now seems to have taken a life of its own. Can they snap out of it? Maybe, but if they don’t do it soon, the season is finished.

    • jessecuster44

      The first paragraph of your response nails the thinking of me, and probably many Reds fans. AVG with RISP may be a flawed stat. However the Reds offense is so terrible that runs are a premium. So when Todd, or Jay, or Brandon, or whoever – can’t get the runner home from 3rd… We harp on it because it’s a blown chance.

      And the fact is the Reds haven’t been doing a good job of getting these runners home for quite some time. Hence the frustration – even if AVG with RISP is a flawed stat.

      And honestly, Thom, or Marty or Cowboy – they should be berating management’s failure, not individual players’ failure. If Walt/Bob had any idea of how to construct a roster, we wouldn’t worry so much about RISP.

    • gaffer

      Agree with your assessment that RISP tells us what has happened, clearly. And the Reds could have won 5 more games if they had hot 50 points higher with RISP. But, that in no way predicts FUTURE performance. In fact, this analysis says that the low RISP will likely increase as it trends toward the mean. All of these same players had much higher RISP in past years and will likely again.

      • greenmtred

        OBP does not predict future performance. No stat does. Stats have some use in identifying players who have a history of performing well, but they do not predict the future. That’s why, for a little while longer, they still have to play the games.

      • jdx19

        Not true. Some stats do a better job at ‘predicting the future’ than others.

        But, if your threshold for predicting the future is exact precision, i can see why you’d say stats don’t predict the future.

        Take walk rate, for example. I will bet $1,000 to anyone who wants that next year, if Votto is healthy and plays a full year, that his walk rate will not be below 10%. This is because walk rate has been shown to be correlated well with itself from year to year, and we have enough of a sample of Votto to be reasonably sure that it’s accurate. AVG, however, is very different. I would NOT make the same bet with average, even if you gave me under .250 or something.

      • greenmtred

        JDX19: I really appreciate your intelligent responses, and certainly agree that some stats would be more predictive than others. My point, which you got, is that no stat is precisely predictive in a given situation. My general opinion is that stats are great tools for evaluating players, making roster and trade decisions and understanding what actually happens and why. And a team does well to pay attention to them in making in-game decisions (whom do you pitch when you really need a strikeout). All that said, though, Chapman might still give up a walk-off home run. So they still have to play the games, and surprising things still happen. Fortunately.

  3. Vanessa Galagnara

    yawn. You guys discredit RISP and Batting average and yet… as a team they show just how terrible the Reds offense is. Because we love certain players we try to cover up the flaws with advanced metrics to show us how great they are…. and yet we don’t score any runs.

    It is great to hear how hard a guy is hitting the ball, how fantastic OBP is, how incredible a W to K rate is and yet I can also say those are flawed stats as well because we are not scoring any runs.

    I measure good by scoring a run not by getting on base.

    • gaffer

      If that were true then players would have stable avg with RISP over their carreers and great players would be better than not so good players. But this is not the case. Look at any player over a long period of time and their avg with RISP usually ends up being the same as their average. Yes, there are years it is better or worse but that is just statistical variations, which you cannot plan for. Statistics at least are somewhat predictive of future performance where RISP avg is not. See Craig, Allen as an example.

      As for not caring about getting on base, how can you have a high average with RISP with no one on base? You clearly need both but you can get on base by hitting or walking and can drive in runs by either also.

      • Vanessa Galagnara

        Not diagreeing but there is some merit to,RISP as a team stat. Batting average sorry I still like it. Shows how good a batter is at getting hits…. Has always been a top measurement of batters and should always be. Number of hits per atbat,,, how can that be wrong? Sure there is something to OBP and maybe it is better but Im still not convinced that walking helps as much as getting a base knock.

        One of my biggest peeves in baseball stats is when a guy gets credit for an rbi by walking…. What was batted in? The b in rbi stands for batted right, and yet the batter did not bat the ball.

        Walks and walkers can certainly help a team and not to discredit any players but fans go to see the studs of baseball hit, not to walk. Got that off my chest whew!

      • gaffer

        I think there was a “statisical” analysis that was done that calculated the “runs” associated with each potential outcome based on the last 80 years of baseball. It roughly showed that a hit created abot 0.5 runs, a double 0.75, a triple around 1 run and a HR about 1.5 runs. I believe a walk was about 0.4 runs. So, yes a walk is not as good as a hit but it is not far off. Its not that walks should be overvalued, but think of how many ways you can score without a hit, but you must get on base first. So, if you walk 100 times in a year (40 runs) and you hit 20 HR (30 runs) then you will be scoring more runs than 50 walks (20 runs) and 30 HR (45 runs).

      • Steve Mancuso

        Those values are calculated by sabermetricians every year based on actual baseball outcomes. How much a walk contributed to runs scored, a single etc. No one *no one* says that a walk is as good as a hit. But it’s a lot better than nothing.

        In 2015 (again, based on actual real games):

        A walk is worth .690 run.
        A single is worth .887 run.
        A double is worth 1.267 run.

        So a walk is not worth as much as a single (obviously) but it’s pretty darn close. That’s why looking at a batters OBP (walks plus hits) is more meaningful than looking at just their AVG (hits). Walks contribute a LOT to scoring runs.

        Here is the research breaking it down year by year since 1871. http://www.fangraphs.com/guts.aspx?type=cn

      • tct

        There is not a single player in baseball who goes to the plate trying to walk, though. Walks are the result of a good eye and a disciplined approach. You are not going to get a good pitch to hit in every at bat. So the best hitters will take a walk instead of chasing pitches and striking out or making weak contact. This approach leads to a higher OBP and a higher batting average, not to mention more runs.

        The best hitters in history, Ruth, Williams, Mays, Aaron, Bonds, etc , all walked a lot more than average. Not because they were trying to, but because they refused to get themselves out on pitches that they can’t drive.

      • Steve Mancuso

        Joe Morgan walked 20 percent of the time in 1975 and 1976 when he was batting third for the Big Red Machine. He won the NL MVP both years. Higher career walk rate than Joey Votto.

      • gaffer

        You could go further and say that the REASON the best hitters ever walked a lot is that if they had swung at those pitches they would have hit a lower average and then we would not be saying they were the best hitters ever. Cozart would probably be a .280 hitter if he walked 14% instead of 4% of the time. I am not sure why this discussion always has to be about walks. Walks are not the issue, but they tell a larger story about the quality of the hitter. Heck, even if walks just lead to the opposing starting pitcher thowing 1 less inning per start THAT has a clear outcome advantage! I get tired of seeing mediocre pitchers get into the 7th inning throwing 80 pitches against us.

      • docmike

        As flawed as the RBI stat is, why on earth would a batter not get credit for a bases-loaded walk? If anybody should love that result it’s you, with your fixation on scoring runs as a measure of a batter’s success.

        As for your other point, no one is saying that a walk is a good as a hit. In some situations it can be, such as a walk or a single leading off an inning. And with a runner on first, a single and a walk almost always still lead to the same outcome, runners on 1st and 2nd. Obviously, with runners on 2nd and 3rd a hit will have a better outcome than a walk.

        But where your argument falls off is in assuming the situation is only between a hit or walk. Like, if a batter got a walk, then that means he missed a chance to get a hit. But there is a third outcome, that being an out. Something even the best hitters do 2 out of every 3 at-bats. And a walk is MUCH better than an out.

        This is why it’s a silly argument when people want the Reds’ hitters to “expand the strike zone” with runners on. If Votto comes up with runners on 2nd and 3rd, and the pitcher decides he is not throwing a strike, would you really want him swinging at pitches out of the zone? Because I can tell you what the most likely outcome is, and it’s not a run-scoring hit.

      • Matt

        Steve, one question on the Fangraphs link you posted – it has the value of a walk as .690 this season and the value of a HBP as .722. How can the value of a HBP be higher than the value of a walk? Don’t they produce the exact same outcome?

      • Steve Mancuso

        That’s always been an interesting mystery. (Keep in mind, these aren’t theoretical. They are based on the actual games.) The one theory I’ve read, and I haven’t carefully studied it, is that pitchers get more flustered by the HBP than by the walk. It’s also possible that walks help pitchers navigate lineups a bit whereas HBP are usually not so calibrated.

      • jdx19

        Hm. I hadn’t noticed that HBP discrepancy before. Very interesting.

      • Steve Schoenbaechler

        “A walk is worth .690 run.
        A single is worth .887 run.

        So a walk is not worth as much as a single (obviously) but it’s pretty darn close”

        This may be a matter of perspective. For, in absolute terms, yes, a .197 difference may not mean much when talking about # of runs. But, then, when you consider how many runs are scored each season in MLB, that number of runs can be huge. Or, if you consider the %change from a walk to a single and how this stat deals with not just runs but actual run production, we are talking about a 29% increase in run production between the two aspects. That can be significant, even moreso when you consider there are more singles hit than walks during a season.

      • Steve Schoenbaechler

        Matt, “one question on the Fangraphs link you posted – it has the value of a walk as .690 this season and the value of a HBP as .722. How can the value of a HBP be higher than the value of a walk? Don’t they produce the exact same outcome?”

        That’s where advanced metrics fall; just like all statistics, they don’t explain why or where. One possibility could be, for instance this. I would think the only way a walk or HBP could score a run would be with the bases loaded. If that is true, then apparently, more there would be more HBP with bases loaded than there are walks.

    • Kurt Frost

      Yet you can’t score a run without getting on base. Until the Reds as an organization start valuing players who can actually tell a ball from a strike the offense is going to stink.

      • MrRed

        Not entirely true. You can rely on home runs. Which the Reds have seemed to do. But swinging for the fences doesn’t produce very consistent offense as we have seen.

        In any event, a lot of the “debate” over statistical measures is more about the rhetoric and language being used and not the underlying merit of the arguments being made.

      • jdx19

        Spot on. Most of our arguments seem to be based on a misunderstanding of the underlying words being used, rather than the concepts the words are tied to.

    • HerpyDerp

      Do you understand how statistics actually works? The guys doing the analysis have literally millions of ABs to determine trends (somewhere around 30 million). With that amount of data points, the advanced metrics become robust. They are more relevant and more accurate than any traditional stat, period. You can say they are flawed, but over the long haul, they are lightyears ahead for predicting outcomes. All I’m hearing is eye-test and ignorance.

      • wkuchad

        yeah, I echo my comments from above

      • Robby20

        Is arrogance a positive quality?

        The bottom line, the Reds stink and it just so happens they don’t hit well with runners in scoring position. Coincidence? Not entirely. Clearly the Reds need to dramatically improve the number of opportunities to bat with runners on base. Meaning they need more base runners. As pointed out the GM says he gets that yet does nothing to address it.

      • HerpyDerp

        They do stink, but taking one partial season for one team and discounting advanced stats based on that is beyond absurd.

      • Steve Schoenbaechler

        You still have to look at the game. Advanced metrics aren’t necessarily some panacea. If the game was that easy, then there would be no need to play the game. We would just play all the games on the papers of advanced metrics.

        Advanced metrics are still only metrics. They never tell why. They are never 100% correlated to tell something. Any sabermetrician worth their weight would be able to tell you that. Anyone who tells you different don’t truly know their sabermetrics. They are putting themselves on a pedestal that just isn’t there.

        Like I said before, schools in figuring out what group of kids to focus on the most, who probably have more data each year that MLB has in AB’s in entirety (aka their data would be just as robust), have determined that, to get the “biggest boost” in their school “performance”, they should be concentrating on the “left handed, hispanic females in wheelchairs who come from single parent (father only) homes and living in poverty”. Digging deeper, getting the “more advanced statistic”, that’s what the statistics showed. But, it just so happens that there is only one student of those throughout the entire school system. Thus, the that statistical would be exactly false to focus on. That, odds are, one of the groups identified earlier in the study would provide the biggest boost to the school, like possibly the females or the special ed kids. Like here, the advanced metrics were essentially entirely needless.

      • Steve Schoenbaechler

        Not that I consider advanced metrics meaningless. I have said on here I would consider advanced metrics in making a team, but not solely on that. There is always the “human factor” that can never be read from a statistic, no matter who thinks they can. For example, it’s been said by some that Johnny Gomes provided a big boost to the Red Sox in one of their WS titles. Why? He helped loosen up the players and locker room. Try to find a statistic on that. Or, Scott Rolen was said to have provided a sense of player leadership that was needed for that team. Try putting a statistic on this. But, every person in baseball will tell you that these can be invaluable characteristics to help build a successful team.

      • jdx19

        Not sure I follow your logic. The basic point of the article is that batting average is poor way to judge performance, especially when you start slicing it down into smaller and smaller samples (like RISP; or man on 3rd, <2out). So, it would seem that your example of the disabled Latina actually supports the premise of the article, rather than showing something poor about 'advanced metrics.'

      • Steve Schoenbaechler

        Actually, JDX, it doesn’t. As I specified, there would be only one student like that. So, even though that one student would provide a bigger improvement to the school score than an improvement from any other student, solely concentrating on that sole student wouldn’t give the school the biggest improvement, period. The school went so deep, getting into the advanced metrics, that they got into minutia. They needed to stop with a more generic group and not delve into the advanced metrics so much. Concentrate on improvement of a bigger group of students, you get a bigger improvement of your score.

        Pretty simple, actually. Putting it in baseball terms for you, possibly, advance metrics could show that the Reds have a problem with having no right-handed Canadian starting pitchers with half a left arm who regularly throw the ball over 100+ mph and also has a Niekro-like knuckleball. And, if we had a pitcher like that, that would greatly improve our ballclub. So, what are we going to do, go seek out a pitcher that fits that mode? Of course not. Even if it would greatly improve our ballclub, it would be a ridiculous waste of resources. We should concentrate instead on getting starting pitchers who throw 95+ regularly. That would provide us much more of a pool to pick from as well as uses our resources more wisely. Or, we get a bigger boost to our club.

    • Kurt Frost

      It’s not that the Reds can’t hit with RISP, it’s that they can’t hit AT ALL!!

      • gaffer

        Agree, but when they occasionally do hit it would be nice to have someone on base. A solo HR could become a multi-run HR.

  4. IndyRedMan

    I knew we had to be bottom 5 atleast. In addition to that I would wager we’re bottom half or worse in allowing 2 out rbi hits. We have to be atleast bottom 10 in scoring men from 3rd < 2outs as well. Votto hasn't even got it done lately? If the Reds didn't play in GABP we'd be bottom 3-4 in runs

    • jdx19

      GABP doesn’t have much effect on run scoring. It increases HRs, but decreases singles, doubles, and triples due to the smaller dimensions of the field. This is backed up by the “park factor’ adjustment used in any stat with a “+” attached to it.

  5. WVRedlegs

    Thanks Nick for some perspective on this matter. Thanks for the information. You guys do good work in educating the masses, which I am one of.

  6. Art Wayne Austin

    The psychological make-up of a player is a factor in evaluating a prospect. Baseball character is lacking on our current team. RISP is a good measuring stick of total immersion and dedication a hitter has to bring to the game of baseball. A player like Matt Carpenter scores high in this regard while Bruce scores low. Was it evident when these players were recruited? Sure, the Cardinals are known for their fiery players(type A personalities) while the Reds are known for being good guys(type B personalities). It’s great to get along well over 162 game schedule but it depressing to be a doormat for the Type A teams. Hello cellar.

    • gaffer

      When a team is winning, Bruce is a perfect fit on a “laid back team” who “doesnt get riled”. On a losing team, Carpenter is a “manic player” who is “presssing”. All this “psychology” is relative to the biases of the outcome. These same players won the division 3 times! Stats actually DO predict future performance over a longer time frame *just ask Wall Street, Vegas, NASA, the national weather service. . .. .

      • greenmtred

        I wouldn’t ask the weather service, if I were you. Predictions of future performance may be correct, generally, but not necessarily so in a specific situation.

    • Steve Mancuso

      Total speculation. No research or other foundation for any of this. Bruce’s career stats are about identical for RISP compared to no one on base. Have you ever met Jay Bruce or Matt Carpenter? What’s your basis of judging their “psychological make-up”? Is this something you can tell from watching on TV? If so, please fill us in on your diagnoses for the rest of the players.

      Research has shown that over time RISP converges to the hitters general batting average. That’s not an opinion. Given that *fact* please explain your assertion that “RISP is a good measuring stick of total immersion and dedication a hitter has to bring to the game of baseball.”

    • Steve Mancuso

      Certainly players have different levels of “immersion and dedication” but those show up all the time, not just when runners are on base. Are you saying Matt Carpenter or Joey Votto tries harder when runners are in scoring position, but not when the runner is at first base, or when no runners are on base in a tie game? That’s ridiculous. Votto is a better hitter overall than Zack Cozart. Whether runners are on base or not. Maybe it’s eye-hand coordination, maybe it’s how much video he watches, maybe it’s “immersion.” But the research shows that over time, those qualities show up equally when runners are on base compared to when they aren’t. And that’s the way it should be.

      By the time players get to MLB, they’ve pretty much proven they can handle stress and are dedicated to their craft.

      • greenmtred

        It’s not a question of trying harder, Steve. It’s a question of being calm and maintaining focus in stressful situations. Not everybody does that. Some people are very capable in tranquil situations, but not when the s### hits the fan. I’m not speaking of baseball here, specifically, but given that ballplayers are still humans, I expect it to be true of them. MLB guys are probably generally pretty good at performing under fire (see, I think I anticipated your answer), but there would still be variation. Interesting discussion.

      • jdx19

        So, 40,000 screaming fans while batting against Randy Johnson is ‘tranquil,’ but 40,000 screaming fans while batting against Randy Johson with a man on 3rd base is ‘stressful?’ That’s about as thin of an argument as you can get.

        The prevailing thought that I’ve seen is that ANY pro athlete has to have a baseline solid mental makeup to even get this far in sports. You can’t going to find a guy who is nervous while batting who is also a professional in the major leagues. I understand there are “head cases,” but that’s not what we’re talking about here. We’re talking about regular players who have excelled at every level from little league to the minors, and now all of a sudden they are nervous?

      • greenmtred

        Not a thin argument at all, JDX19. 40,000 screaming fans wouldn’t be a tranquil situation, but, besides being highly unusual, would be less stressful than 40,000 screaming fans with the winning run on 3rd. Most people, pro athletes included, can handle varying levels of stress. As situations get more stressful, however, fewer people are able to handle them well.

    • docmike

      That’s a ridiculous argument. A good hitter is a good hitter, whether there are RISP or not. And a bad hitter is a bad hitter in either situation.

  7. georgestricker2

    Last night’s game proves that this conclusion is wrong. Had Reds score with Cozart followed by Pena at the plate complexion of game would have changed dramatically. Statistical analysis is certainly important but not determative of a team’s success. RISP does affective a team’s in game strategy and thus the game’s outcome. You cannot analyze a team’s success or failure by statistics which seems to be the premise here. Other more subjective factors are just as significant.

  8. George Mirones

    Nick; Great job of analysis. Time and effort is appreciated.
    Overall my take of the entire article can be summed up in your two insightful and accurate facts;
    “The Reds’ problem scoring runs the last several years has been due to overall mediocrity in all facets of the offensive game, not due to a glaring inefficiency with runners in scoring position.”
    And
    “Think about it, which pitchers allow more baserunners? Good pitchers do not allow you to get a runner into scoring position very often, bad pitchers do.”
    The often ignored stat, “WHIP”, can say a lot about the pitchers overall skill in not allowing base runners;
    In baseball statistics, walks plus hits per inning pitched (WHIP) is a sabermetric measurement of the number of baserunners a pitcher has allowed per inning pitched. Since WHIP reflects a pitcher’s propensity for allowing batters to reach base, a lower WHIP indicates better performance.
    Cueto
    2013 – .88
    2014 – .96
    2015 – .96
    J. Marquis
    2013 – .152 Padres
    2014 – 1.26 AAA
    2015 – 1.63 Reds
    These numbers tell the story we all know and love!
    This may seem like an overly simplistic approach but my questions would be; How do the Reds perform against pitchers with a whip of plus 1.0 versus against pitchers with a whip of minus 1.0.
    Basically my idea is that if the Reds perform below average (get on base less) against substandard pitching (high WHIP) then it is strong support for the “overall mediocrity” of the team.
    The comparison, numbers wise may show that against substandard pitching the Reds are still substandard in hitting while other teams regardless of RISP are just better hitters overall.
    I hope this makes sense.

  9. liptonian

    Can someone do an ask Marty and just send the link to this?

  10. bohdi87

    Great article. Objective, quantifiable evidence. The comments disagreeing make my head spin.

  11. docmike

    I would add one caveat to this wonderful article. I wouldn’t say that hitting with RISP doesn’t matter. Of course it matters, if the Reds were batting 1.000 with RISP we would probably be in first place right now! Statements like that are what get the old-school stats people riled-up.

    What should be gleaned from the stats is that hitting with RISP isn’t a special skill. If you aren’t good with RISP, it probably means you aren’t a good hitter overall. The Reds don’t need to get better with RISP, they need to get better at hitting, period.

    The other important point is that the more opportunities with RISP you have, the more runs you will score. The best teams get lots of runners on base, thus giving them more chances to score runs. That is why OBP is a much more valuable stat than hitting with RISP.

  12. seat101

    Before I read everyone’s comments, I have to say this is one of the best written articles where math is an integral part of the story being told.

    Yet a layman, neither a baseball stat person nor a strong math head can easily understand the concepts and the arguments expressed.

    The fact that I agree with the point of the essay is not coloring my view of the tremendous writing.

  13. Greg Dafler

    I think the main takeaways here are that

    (1) AVG w/RISP rank does not correlate to run scored rank. In other words, a high (let’s say top 5) AVG w/RISP doesn’t make your team a top five offense.

    (2) the Reds batting average with RISP stinks, but so does their overall batting average. When one of the radio/TV guys gets on the Reds AVG w/RISP, I wonder don’t they realize the Reds AVG isn’t all that great either?

  14. Tom Reed

    Will Marquis be able to put a stop to the losing streak?

    • Tom Reed

      I commented on the wrong post. Should have been ‘Titanic Struggle Recap….’

  15. beelicker

    Ignores that the Reds are STILL an aberrant outlier in team DOUBLES, plus STILL the only team in MLB whose HR total is equal to (which it still is) or not less than doubles + triples. Sure small sample size, but this Reds team in FREAKISHLY deficient in 2B knocks: http://espn.go.com/mlb/stats/team/_/stat/batting/sort/doubles/order/true

    More outs are obviously spent in trying to move runners ISP from 1st than if they put themselves ISP via XBH. Given their tepid OBP & if not for the apparently prevailing tendency to view themselves ISP in the batter’s box (#4 MLB in HRs) this team would be seriously run-challenged even more than they already obviously are: http://espn.go.com/mlb/stats/team/_/stat/batting/sort/homeRuns/order/true

    • jdx19

      The dearth of doubles has been very odd. I think it might have to do with our HRs being relatively high. Already this year, off the top of my head I can think of 5 first-row homers that have been hit and could easily have been doubles.

      I think the fact that the team is a poor hitting team to begin with, and that a lot of would-be doubles have become HRs is contributing to the weird lack of doubles.

      • beelicker

        Last night was something of a season watershed (& still subject to going back) in that 2 2Bs vs o HRs FINALLY lifted them out of having more total 2B + 3B than HRs

        Boston (Green Monster limits doubles vs cheap HRs) , Baltimore & Houston (mostly because of their rocketous HR total) are the DH league teams that come ‘closest’ to an underwater ‘other’ XBH/HR ratio like the Reds

      • Shchi Cossack

        The Reds as a team rank 13th of the 15 NL teams with a .238 AVG, but rank 8th of the 15 NL teams in SLG. This is exactly as WJ built this team. Once again, it’s not so much a problem of AVG with RISP, this is just a poor hitting team that relies almost exclusively on the HR with high SO%, an all or nothing approach.

  16. Nick Kirby

    Great article, and a very important one with the ignorance coming from the Reds broadcasters. Thanks for this piece Nick!

  17. Mary Lykins

    This RISP stat is one the stats fanboys can’t grasp. There are players who are “clutch” and there are players who are not. There are players who relish the chance to be in those pressure situations and there are those who shy away.

    • beelicker

      One way to try to help ‘prove’ this would be to separate & compare PAs w/ RISP & 1st base also occupied vs ISP w/ 1st base open

    • Nick Doran

      Who are the clutch players? Throw out some names and we will see if your theory that they are clutch is supported by their stats.

  18. Brian Dunn (@surrfinTexas)

    Nick (and Steve), great pinpoint subject on our little Reds. So weak in April and now May, with RISP. First inning tonight, bases loaded 1 out, I check the score 15 minutes later, 0-0. #letsGoReds

  19. Matt

    Bottom line is this team is woeful offensively at the moment. They have scored the majority of runs via the home run. You cannot count on the 3 run homer in today’s game. You have to get them on, get them over, and get them in. The inability, outside of Bryan Pena and Votto, to shorten swing and hit the ball to the opposite field with a 2 strike count is inexcusable. Reds have been a poor bunting team overall as well. When does Jay Bruce learn you can’t pull the ball and expect solid contact when everyone pitches him low and outside. It is beyond me to comprehend. And the unwillingness to push a bunt towards third against the “shift” when you are struggling is just plain ignorant.

    I like Bryan Price and I think he is a good baseball guy, but I think he will be the one to be sacrificed if the current slide continues. Something has to be done to wake some people up.

    When do you give some younger guys a chance? Why hasn’t Lorenzen/Iglesias taken over Marquis’ spot permanently? Great guy, great character but has no “out’ pitch any longer and should be release.

    Watching this team is like watching ants crawl.

    • jdx19

      Might be out of line to call Bruce ‘ignorant’ for not pushing bunts down the 3B line. I, to be honest, don’t think he has the skill to do that because he hasn’t practiced it enough and does not have coaches capable of teaching him how to do it. Do I agree I wish he would do that someimtes? Yes. I don’t think he can, though. All ballplayers can’t do everything well.

  20. Steve Schoenbaechler

    It looks like to me there is a correlation to runs/game and BA wRISP. It’s just not a 100% correlation. But, there definitely seems to be a correlation. Probably more visible on an actual graph rather than a table, it seems to me that most of the better teams with high BA wRISP have higher runs/game. Factors that could throw that off: possibly what hits are actually hit (a single wouldn’t drive in as many runs as a HR), how many runners on base (2 run HR compared to a grand slam), how many runs were scored when there weren’t runners in scoring position.

    But, does it matter? I would think absolutely yes. For instance, the 3 division leaders in the NL are all in the upper half of BA wRISP. Again, will there be a 100% correlation? Most likely not. But, does it make a difference? You might ask yourself this, would you rather have your team with a high or low BA wRISP? I would rather have a high BA wRISP.

    • jdx19

      There certainly is a correlation, but how strong is that correlation? The fact that a correlation exists or doesn’t exist is useless in isolation. For the team data presented above, R-Squared between AVG with RISP and R/G is .2377. That isn’t a terribly strong correlation.

      No one has ever said “it doesn’t matter.” Even the article does not state that. It’s just a very poor way of evaulating teams and players, given the other ways that exist of evaluating teams and players.

      • Steve Schoenbaechler

        “It’s just a very poor way of evaulating teams and players,”

        That’s your opinion. Personally and professionally, I wouldn’t mind at all having some players who can come through when some pressure is on, like when runners are in scoring position.

      • jdx19

        It’s fact. Sorry. Just because you don’t want to believe or understand it does not make it an opinion.

      • Steve Schoenbaechler

        As I just showed below, you seem to be basing your opinion off faulty information.

      • Steve Schoenbaechler

        Sorry you don’t accept it.

      • Nick Doran

        I think you missed the point of the article. If you want to know how likely a hitter is to get a hit with runners in scoring position you should not look at his batting average with runners in scoring position, you should look at his batting average. AVG w/RISP is nothing more than a small sample mirage like AVG on Saturday. This has been proven over and over by people a lot smarter than me. AVG w/RISP is not a distinct talent or skill. Good statisticians can easily see that.

  21. Steve Schoenbaechler

    I am still a person who believes that “situational conditions” can still be practiced. It has been said by Michael Jordan himself when asked how is he so good in the “last second game winning shots”. He answered, “I have practiced those for years.”

    Athletes can freeze up given certain conditions. That isn’t uncommon nor abnormal. For every player? Of course not. So, will RISP be similar to regular BA? Sure, but again, that isn’t a 100% correlation, because of the condition. That’s where the mental makeup of the players, the human factor, comes into play and cannot be quantified. The closest thing to it that I have possibly seen, like in this case, where a BA may be high but the BA wRISP may be low. And, until those who don’t think this aspect exist starts to accept this aberration or that there is a “human factor that can’t be quantified”, then it is pretty useless to discuss items like this with them.

    • Nick Doran

      Which players fit your criteria? Examples.

  22. jdx19

    Did a quick-and-dirty spreadsheet about how each team stat correlates to runs scored per game through last night.

    R/G to R/G: 1.00 (Good. Our math works)
    R/G to Contact %: .0075 (Threw in something weird to check for small correlation)
    R/G to BABIP: .155
    R/G to AVGRISP: .257
    R/G to ISO: .271
    R/G to AVG: .310
    R/G to OBP: .446
    R/G to wRC+: .561
    R/G to SLG: .565
    R/G to wOBA: .615
    R/G to RBI: .883 (Another sanity check, you’d assume this would be close to 1)

    Soo, yeah. AVG with RISP is a poor measure, as the author states, to base evaulation of run scoring against. It’s even worse that regular old batting average.

    • Steve Schoenbaechler

      Run the same on the Reds themselves.

      Personally, I wouldn’t mind at all having a player(s) who can come through when the pressure is on, aka when there are runners in scoring position. And, the fact that the Reds can’t get them in very well is still a weakness of this team right now.

      Oh, in general, it might be a poor correlation. But, that doesn’t mean at all that for a specific team that it’s a poor correlation. As well as, the stats don’t speak of conditions for a game. For instance, going up against someone like Bumgarner on the Giants or Kershaw on the Dodgers, when you get a guy on 2nd or 3rd, you better be able to get him in. Because, odds are, if you don’t, you aren’t going to win that game. Against a Marquis-like pitcher on other teams, I wouldn’t care about any runners in scoring positions. We have won a game like that hitting only HR’s. But, against the best of the best, we have to be able to take care of business when we can. Pitchers like Bumgarner and Kershaw, they don’t make many mistake pitches giving up HR’s. So, we can’t rely on those. So, we better be able to hit with RISP. And, if we can’t do it against the weaker teams, we sure aren’t going to be able to do it against those pitchers.

    • Steve Schoenbaechler

      Also, you said yourself, “how each team stat correlates to runs scored per game through last night”. That’s the analysis you ran. Then, you state:

      “Even the article does not state that. It’s just a very poor way of evaulating teams and players”

      I suggest run an analysis like this on players. But, wait, do you do “runs scored” per game? or, “runs batted in” per game? Not necessarily the same statistic, including the one you ran, which you specified was for “each team”. It sounds like you are comparing entirely different items here.

    • Steve Schoenbaechler

      Not sure how you did your math. I just ran a correlation on the table, and I get an “r” value of 0.45 for R/G to BA wRISP, using the place the team finished in the table. Using the #runs/game, I get an “r” value of 0.49, almost twice as much as you got.

      • jdx19

        R-squared is what I was using. Do some reading on correlation coefficient versus coefficient of determination. Both are somewhat valid in most situations, but mean slightly different things.

      • jdx19

        Put in another way, take the square root of the numbers I posted and you get the “r” you are referring to.

      • Steve Schoenbaechler

        Actually, you don’t, JDX. Please do the math.

      • jdx19

        I’m not going t0 take the time to walk you through it, dude. Nothing is wrong with my math or my statement.

    • Nick Doran

      Thank you JDX19, your numbers are very conclusive that AVG w/RISP is one of the worst possible ways to evaluate hitters. Many people will never admit it though. It is tough to admit one is wrong after years of arguing so vociferously for a losing cause.

  23. Vanessa Galagnara

    Stats can be used to slant any argument. Just ask a politician. There is no point in doing a statistical analysis to tell us this player is good or bad with RISP because the only stat that matters for hitters is how many runs come across that plate. Do they drive in runs, do the add a positive element to the ability to score runs? This season Todd Frazier comes to mind. Where would we be without Todd Frazier?
    A few other fellas that are M.I.A. and yes they are not driving in runs when the Reds need them to. No stats. Just look at that win and loss record and I dare you to defend your favorite players. Only Todd is performing as expected. Say anything else and you would lose all credibility.

  24. Darren

    Does it matter?/ You wrote an entire article about something that is as obvious as you can get. YES it matters. DUH!

  25. Tom

    Great article. The Angels announcers do the same thing. If they go 0 for 6 with RISP they make it sound like it’s an incredible trend to be discussed in detail.

    What I want to know is, what percentage of total runs scored in baseball are scored by runners in scoring position? I know it’s a mouthful. I would guess it’s not much over 50%, further demonstrating the silliness of this statistic.

    Tom