It doesn’t add up

In today’s Detroit News Tigers beat writer Tom Gage writes about RISP. RISP of course is runners in scoring position and Gage goes on to explain part of the Tigers woes by looking at their performance in RISP situations. Along the way he creates a new metric and applies some faulty logic.

Gage notes that the Tigers are the only team in the AL Central with a RISP batting average lower than their overall batting average. Gage then goes on to compare the Tigers to the division leading Twins:

The Tigers are hitting .264 as a team but .251 with RISP. That’s a difference, math majors, of minus 13. The division-leading Twins? After Sunday’s game, they were at plus 46. Their overall batting average was .265, but with RISP, they were hitting .311.

But here’s the kicker, the Tigers have outscored the Twins this year despite the disparity. The Twins have a better record because they are allowing a run a game less than the Tigers, not because they have a better RISP batting average differential.

Next Gage goes on to talk about a new stat loosely termed combined average. It is described thusly

When you combine the two numbers just like combining slugging average and on-base percentage gives you a better perspective about production, you get a total average that provides a more complete view of a team’s offense.

The Tigers’ total average is .515. The Twins’ total average is .576.

That’s also why the Tigers’ apparent offensive edge over the Indians is misleading. The Indians are hitting .243 as a team, 21 points lower than the Tigers. But they’re also hitting .287 with RISP for a total batting average of .530.

First of all I’m not sure why this would provide a more complete view of production, but I’ll be back to that in a minute. Once again the Tigers are compared to a team that has been their inferior in terms of scoring runs. The Indians score 4.47 RPG where the Tigers score 4.74.

But as for the assertion that it provides a more complete view of production, it just ain’t true. Probably the most complete view of production is runs scored, but of course we want to better understand the whys and hows of run scoring. So as we look for answers it is probably worth looking at how any new stat actually relates to run scoring.

Using AL data for this year (which is what Gage is using) I looked at how combined average related to runs. Combined average had a correlation of .43. (Correlation values run from -1 to 1. A value of 0 shows that two items aren’t really related. A value closer to 1 or -1 shows they are highly related.) Now .43 isn’t bad, but we can do better.

Plain old batting average has a correlation of .60. On base percentage and slugging percentage are .79 and .77 respectively. And OPS rocks out a .83.

Looking at RISP stats to see if they are more highly correlated with run production than overall stats we see that RISP batting average checks in at a meager .21 and RISP OPS .39. You know what RISP stat does correlate with runs scored? At-bats with runners in scoring position has a .66 correlation. So one could say the number of the number of RISP opportunities has more to do with run production than the actual performance in those at-bats.

The problem with Gage’s stat is that you’re adding things that are kind of alike, but they don’t get weighted by opportunity. Plus, the RISP performance is factored into the overall already. The other problem of course is that a quarter of the way into the season, numbers are still pretty volatile, especially when looking at splits like RISP where there are only 300-400 at-bats.

There is some value in looking at RISP performance, like if you wanted to see if there was a disparity between a run model like runs created and actual runs scored. But otherwise it’s more of a novelty. (As an aside, using the short form of Runs Created=TB*OBP the Tigers are only 3 runs short of their estimate.)

The Tigers offense, while amassing decent totals, has been a problem due to its inconsistency. In the 22 losses their OBP and their slugging percentage are less than .300. But number like that speak to bigger problems than just their inability to hit with runners in scoring position.

77 thoughts on “It doesn’t add up”

  1. Billfer – if its not too complicated, any chance of explaining how you get the correlation number?


  2. Here’s a link to explain correlation.

    Looks like Billfer is taking the all the AL team batting stats and correlating them with the number of runs that team has scored.

    A simple correlation example: Suppose 3 teams have batting averages of .250, .280, and .300. If those teams have scored 125, 140, and 150 runs, then average and runs are perfectly correlated (1.0) since runs = average / 2. If those teams had scored 200, 150, and 120 runs then there would be a negative correlation between average and runs since higher average = less runs.

  3. Wow. Batting average plus batting average with RISP? Really?

    When I first read about OPS, I didn’t like it because it just felt like a number. Batting average is hits divided by at bats. On-base percentage is times on base divided by plate appearances, etc. OPS’ equation was simple enough, but it was made up of other stats so when a player did something on the field, the impact on his OPS wasn’t as intuitive to me. Actually, this slight complexity compared to other stats is probably why it (along with other more advanced stats) meets some resistance.

    Gage’s ridiculous stat not only has the feel it’s not measuring something; it instantly feels completely useless. Why stop at BA w/RISP? Why not add a bunch of other situational batting averages and call it something like “airtight totally complete no hidden trickiness average”? Just keep throwing them in there until they tell you precisely what you wanted them to before you started the pointless exercise.

    I mainly quit dogging on the mainstream media quite a while ago, but this is just ridiculous. I hope for Gage’s sake Fire Joe Morgan doesn’t get a hold of this.

  4. Interesting. Yet another reason to love this blog. Respect for facts, examination of same. Not merely people trying to outdo each other in shouting their opinion the loudest. But we knew that.

    Something that is subjective is the question of what has been the biggest disappointment so far this season. For me, it’s the hitting. The Tigers aren’t hitting very well with runners on. I don’t know far below the league average they are in this respect, but I think it’s reasonable to expect that a world-beating lineup could do better than a 13-12 record in games where the opponent scores 5 runs or less. Or not get shut out 5 times in 38 games.

    It could be worth looking into how consistency of scoring correlates to winning. Consistency of runs allowed, too. The first 18 games, the mean (or is it median?) score was Opponent 5, Tigers 3, good for a 6-12 record. The next 18, it was Tigers 6, Opponent 5, still good for no better than 9-9. That means something, and I think it has to do with consistency.

    Not that I don’t think the lack of quality starts has been appalling, but somehow I remain more disappointed with the hitting.

  5. I don’t think Gage is a particularly good writer – I gave up on him a long time ago. Besides, why would anybody bother with that mush when DTWB is available?

  6. I think a more useful analysis than looking at averages is to look at some measure of variance in the team’s run scoring compared to other teams in the central. My armchair perception thinking back over this season’s games is that we either score a bunch of runs in a game or next to none. Such performance could give you decent averages and other measures on runs scored per game, but would show up as a wider variance or standard deviation.

    John Gasaway, who does statistics analysis on college basketball at basketball prospectus, looked at measures of consistency this season. Michigan State showed up as one of the most inconsistent offensive teams in the country – one game they would go off (Indiana at home) and another (Iowa on the road) they would be abysmal offensively. Unfortunatly the Tigers are this year’s baseball equivalent so far, or at least that’s my perception without looking at the stats. If I get a chance I’ll run variance in runs scored this year for every team in the central and report the results when I get them.

    The bright spot for MSU was that they were consistent defensively, so they won some games where their offense was not up to snuff with D. So far this season, the Tigers defensive statistics (both pitching and fielding stats) are near the league bottom. Basically, they’re not doing anything well except scoring runs, and they are inconsistent even on that one bright spot. That pretty much explains why their performance has been poor.

  7. I think that too often this year, pitchers have found ways to shut down the offense of this team at key moments of the game. The Gavin Floyds of the league have dominated this team. Hitting with RISP is the very thing that knocks out a pitcher who is having success, and the Tigers have not been doing that.

    Hitting with RISP is a valid stat that does reflect the clutch ability of a team. And one should expect that BA with RISP should at least be equal to overall team BA. The fact that there is such a disparity with this team reflects that this is a group of chokers that cannot pull off the big moment that wins games.

    Now I don’t know the numbers, but I’ve heard that our starting pitching opponent’s BA with RISP is through the roof. If so, this would explain why they have been so awful.

    It seams as if you can go up and down our entire roster of regular players and you will find that nearly every one of them is struggling mightily. This season is a perfect storm of multiple players having below career average seasons– all in unison.

  8. Billfer — thanks for the regression analysis. The Gage quote on math majors seems especially ironic — maybe we can send him a statistics or econometrics textbook. I think Mark J’s armchair perception seems closer to the truth, even though the sample would still be pretty small.

  9. There is some value in looking at RISP performance, like if you wanted to see if there was a disparity between a run model like runs created and actual runs scored. But otherwise it’s more of a novelty.

    Indeed. And since you basically disproved “clutchness” last year, all his “differential” really measured was luck.

  10. Ah regression analysis. So many scary memories of grad school invoked. Suddenly, I’m back eating ramen noodles, coding data, and watching Sparky Anderson take a mental health leave.

  11. The fact that there is such a disparity with this team reflects that this is a group of chokers that cannot pull off the big moment that wins games.

    BA w/ RISP for 2007

    Ordonez: .429
    Polanco: .364
    Thames: .325
    Inge: .296
    Pudge: .296
    Guillen: .289
    Sheffield: .288
    Granderson: .256

    Cabrera: .378
    Renteria: .331

    One presumes, these players all became “chokers” during the offseason.

  12. Thank you for this, Billfer. I heard Tom Gage on WDFN yesterday afternoon rattling on and on about this and I was pulling out the last few hairs remaining on my head. He was talking as if our starting pitching wasn’t even a factor.

    The Tigers have scored plenty of runs. Yes, the bats have been streaky but I have full confidence that this lineup can and will produce. They are 5th in the AL in runs scored and Team BA, dead last in Team ERA.

    It doesn’t take a statistical genius to figure out the bigger concern.

  13. These guys need to stick to the reporting and leave the analysis to those who understand what they are talking about. Well done.

  14. It’s interesting to see an analysis of Gage’s idea. But I wouldn’t necessarily rip Gage. At least he is willing to think beyond the norm, which probably is some kind of small step forward. He failed, but at least he tried.

  15. Since Mark J raises the MSU basketball team’s record-level inconsistency this past season, I can’t help but plug my own graphical analysis of this phenomenon.


    I’m not sure if inconsistency in baseball has the same meaning as it does in basketball, though. Basketball is a team-oriented sport; if one player struggles, it’s more likely to affect other players’ performance. Baseball is based more on individual performance, so inconsistency may just be a randomness thing.

    Analysis of individual player inconsistency may be more useful. For example, both Granderson and Pudge strike me as players that are either very good or very bad against a given pitcher.

  16. I have to agree with Ken here. It’s nice to see Gage try. Maybe he could call on an expert to help him a bit more, so he could be, you know, not wrong. But it’s better than nothing.

  17. Just for fun, I wondered if there was any way you could juggle the per-game results for runs scored and allowed to produce a better record than 16-22. To my surprise, there was a way to get a 23-10 out of it (throwing out ties). The key seems to be, when you’re going to give up 11-13 runs, don’t waste more than 1 run of your own, but still don’t get shut out. That’s a joke.

    Lining up results evenly, best against best, the Tigers go 5-24. Lining up results unevenly, worst against best, the Tigers go 16-21 (pretty close to the truth, there). Doesn’t prove anything, but there’s some meaning in it. I’m sure of it.

  18. Sean C.

    Were you able to follow Friday’s game online against the Yankees? I would point out they actually won that one, so your streak would still be intact. Hopefully you are able to jump online a bit more often so we can get out of this hole.

  19. And OPS rocks out a .83.

    So the question is, what can be added to OPS to bring it closer to 1? The truth is out there.

    The search is on for the missing particle predicted by Unified Team Batting Theory.

  20. Mark (in Chicago)

    Yes, I followed the last win against the Yankees start to finish, so I’m 4-0. I deliberately skipped the Saturday game. I can’t trust the Tigers to win 2 in a row. My stats are more important than the team, so I’m going to pick my spots carefully.

    This is a free agent year for me – I hope to be signed by another team blog during the off-season. I’m thinking Red Sox. I need to find a team where I can watch/follow more games without too many blemishes on my personal W-L.


  21. Sean: Make sure you negotiate a no-trade clause into your free agent deal, OK? And maybe a couple blogger option years tacked onto the end.

  22. I think Gage should stick to what he’s good at which is reporting rather than analysis.

    The consistency of run scoring is an interesting concept. It wouldn’t be too hard to do. You could use standard deviation or coefficient of variation. I wonder how much that impacts wins and losses? I think an inconsistent team might tend to underperform their pythagoream win estimate due to a few blow out wins and a lot of close losses. In the Tigers case, their actual wins equal their expected wins so I would guess that their inconsistency is not hurting them, at least not yet.

  23. Chris (in Dallas)

    Between you and me – don’t worry, I’m not going anywhere. On the advice of my agent, I’m dropping hints that I might be restless. But I really want to finish my career here at Detroit Tigers Weblog.

    Still, there’s no denying that I haven’t been getting my propers here. While the so-called workhorses who watch every game have put up a dismal 16-22, I – the new kid on the block, the guy who came from nowhere, the “marginal Chad Durbin-type 5th starter” – have been the skid-stopper, putting up wins after losses or losing streaks against the likes of the Red Sox, Blue Jays, and – twice – the hated New York Yankees. Appreciation? Nada.

    I won’t go so far as to say I’m carrying the team, but I am getting a little disheartened by the lackluster play of my teammates. I’m only going to say this once, folks: Watch to win. Don’t stand idly by and expect the next guy to telepathically urge the Tigers to victory. Take some responsibility. I can’t start every game, at my age.

  24. Kyle – good point. An offense has to be at least somewhat consistent to be considered good in my humble opinion. If you’re dominated, shut down more than any other team, and twice a week you bust out with 16-19 runs, and somehow end up leading the league in runs scored, so what. I don’t think that makes you the best offense. It might just mean you make your living by exceling against scrub pitchers, even though you’re owned by most others.

    I’m oversimplifying here for the sake of brevity, but you get the point.

    A better metric would be to set a bar for adequate production in a game, be it Runs scored, OBP, hits, SLG %, whatever it is, and see what % of the time you meet or exceed that production.

    Who has the better offense?

    Team A scores – 7,7,6,6,7,6,8,8,7,7 – 69 total runs
    Team B scores – 2,3,2,2,19,3,23,2,2,17 – 75 total runs

    Team B scores more average runs, but in actuallity, they get shut down more often than not. That shouldn’t be considered good, as more often than not, they’re not good.

    Team A scores fewer average runs, but they perform well all of the time. Not a single staff has figured them out.

    I’d say team A is not only better, they outclass team B.

    Yes, yes, small sample, but you get the point.

    If you set the bar at 6 runs per game. Team A meets that 100% of the time. Team B 30%.

  25. I’m not sure if inconsistency in baseball has the same meaning as it does in basketball, though. Basketball is a team-oriented sport; if one player struggles, it’s more likely to affect other players’ performance.

    Kyle, I agree with first part, disagree with the second. A Michael Jordan or Kobe Bryant can carry a team single-handedly. While dominant pitching performances in baseball often go for naught. Overall, I think baseball is more of a team sport, in part because you can’t just “give the ball” again and again to the player with the hot hand, and also because you can’t cover as easily for weakness at any given position.

  26. A better metric would be to set a bar for adequate production in a game, be it Runs scored, OBP, hits, SLG %, whatever it is, and see what % of the time you meet or exceed that production.

    Excellent idea, Greg.

  27. I agree that consistency is probably a good thing in general. However, the Tigers pythagorean record is the same as their actual record at this point. That’s why I don’t think their inconsistency has hurt them so far. I think it may be because their run prevention has been so consistently bad that they would have lost many of those games where they didn’t score a lot anyway. They aren’t losing a lot of 2-1 and 1-0 games.

    Another thing to keep in mind when analyzing offense this year is that offense is WAY down in the AL this year. Teams averaged 4.9 RPG last year. This year they are at 4.4 which is a huge difference. Offense usually starts picking up in May but so far it hasn’t. So, the Tigers are not the only team struggling with offense this year. Unfortunately, their pitching and defense is not contributing to the league-wide low runs scored numbers.

  28. Strange season. There are probably only two guys at this point who are exceeding expectations. Lopez and Santiago–and they’re both role players.

    Frankly, the way this team was constructed, basically to outhit the opposition and win a bunch of 8-5 games, the biggest story to me is not the struggles of the pitching staff but the inconsistency of the offense. If this offense was performing the way even the most conserative among us expected, we’d be around .500 right now waiting for the pitching to come around.

    The fact the Tigers are among the league leaders in offense does not, however, excuse the offense from blame in this miserable start. They were built not to be among the league leaders in runs but to be one of the top two offenses in the league (with the Yankees)–and by a wide margin. They were built to win the way the Indians did in the 90s, by bludgeoning the opposition. Instead, we get 19 games (half!) where they fail to score at least five runs.

    The pitching is a big problem, yes, but not an entirely surprising one, although the widespread ineffectiveness is a surprise. Long term the offense will probably straighten itself out, assuming Sheffield and Cabrera find their stroke and Guillen can reclaim the consistency that seems to have deserted him after last year’s All-Star break (anyone check his numbers from July-September last year and the last month or so of this year?). And let’s bring up Larish, for god sakes, and DH him. The Tigers desperately need a left-handed power bat

    The pitching will improve–one way or another. The old regression to the mean business.

    But to this point, in terms of what frustrates me more, it’s definitely been the up and down offense.

  29. Gage shouldn’t be employing statistics to make his point, but it doesn’t necessarily mean there isn’t some point there to uncover. And I give him credit for at least trying…

    If I were creating a metric, my guess would be something along these lines:

    Men left on base as a percentage of total men put on base (weighting the bases accordingly):

    Tigers ratio — not so good relative to the other teams. Much higher than Twins. The Tigers are likely better at team OBP (using the same weighting method as above) and put much more men on base than the Twins and much more than many other teams in the league, hence they score more runs by default — but more than that, the ratio of men left on base to total men put on base is probably much higher than the Twins… just a guess. A team can score less runs than another team but still be more productive in driving in runs, given the opportunities, and visa versa.

  30. If you set the benchmark for batting success in a game at 6 runs or more, and pitching success at 5 runs or less, Tiger hitting has succeeded only 42.1% of the time, while the pitching has succeeded 63.2% of the time. Change it to 5 and 4, and it’s hitting 47.4% and pitching 42.1%. Take your pick as to the more reasonable expectation.

    Coming into spring training, there were question marks hanging over 4 of the 5 Tiger starters. There were no such doubts about the lineup, other than perhaps Sheffield. The bats started slow in 2007, too, but that was with Monroe and Inge and minus Cabrera and Renteria. So I am subjectively more alarmed by the hitting – the team that was supposed to go boom hasn’t. I see starting pitching as the “struggling” part, and the hitting as the “collapse” part. I see both of them coming around, but the question is, will they manage to get it together at the same time often enough?

    No, the Tigers haven’t been losing 2-1 or 1-0. They’ve been losing 5-1 or 5-0 instead. The lower the total runs scored, the more likely it was that the Tigers lost.

  31. It’s interesting that the Tigers runs scored total is so close to their runs created estimate even with their problems with RISP. Generally, a team with a very low RISP tends to have a higher RC than runs scored. I think it may be because they are scoring runs in bunches – big innings. So they get a lot of bang for the buck in their big innings which balances off the innings where they fail.

    Still I think the offensive problems pale in comparison to the pitching/defense problems. The offense is a little worse than I expected but the run prevention is much worse than expected. 5.5 runs allowed per game is incredibly bad in an environment where teams are scoring only 4.4 runs per game.

  32. Sean, I see your point about one player not being able to carry a baseball team. My point was this: If the point guard on a basketball team is having a bad shooting night, the defense will adjust packing it in, making it harder for a post player to score.

    In baseball, if Sheffield is struggling at the plate, it’s less clear that has an impact on how other hitters in the lineup perform. The only angles would be: (1) protection for the guy batting ahead of Polanco, (2) it’s easier to hit with runners on because they can’t pitch around you, and (3) psychological impact. Studies have shown (1) and (2) aren’t huge factors. We can speculate on (3) but I think the randomness of a baseball season usually creates what people believe to be psychological issues.

  33. Pitching/defense are intertwined. A large part of the reason the pitchers have looked so mediocre is the completely atrocious defense behind them. It’s a virtual butcher’s shop out there. Look at Robertson. He’s struck out 31 batters and only walked 11 thus far, which indicates he’s pitching fairly well. Until you look at his ERA. This team can not convert ground balls into outs and the pitchers suffer for it.

  34. Sean, the benchmark thing might be a good idea. It’s kind of like quality starts for pitchers. I believe the 5 and 4 benchmarks make a lot more sense than 6 and 5 when the league average is 4.4 runs per game. Giving up 5 runs would have to be regarded as a failure this year. I’d want to see what other teams look like using these benchmarks.

  35. Actually Sean, here’s the Tigers real win-loss record for the low scoring games:

    0 runs — 0-5
    1 run — 0-5
    2 runs — 0-4
    3 runs — 0-4
    4 runs — 0-1
    5 runs — 2-1

    I should really put together the tigers record by wins allowed as well. So I don’t know if that’s the offense’s fault for scoring 3 runs or less 18 times in 38 games, or the pitcher’s fault for not winning a 3-2 or 2-1 game even once. Some combination of both.

  36. Lee, thanks, but give Greg (see earlier post) credit for the benchmark idea. He took it even further as far as game to game performance goes.

    I think you’re right that 5 and 4 are more realistic standards, my disppointment that the Tigers can EVER manage to score less than 6 runs with their lineup notwithstanding. With that in mind, it would seem that the Tigers problem is that they can’t manage to pitch and hit well in the same game often enough. This inability to get it together is scarier than blaming offense or defense. Nothing speaks “losing season” more loudly, with the exception of a Detroit Lions jersey. And look at the dearth of extra-inning and 1-run games. Something spooky about that.

  37. I just took a quick look at how many times teams have scored 5 or more runs in a game:

    Bos 25
    LAA 21
    TB 21
    Det 19
    Chi 19
    Tex 19
    NYA 18
    Bal 17
    Oak 16
    Min 15
    Sea 14
    Tor 13
    Cle 13
    KC 12

  38. Sean C.

    Don’t sell yourself short, you’re more like Armando Galarraga than Chad Durbin right now.

  39. Actually Kurt, I was thinking in terms of total runs, both teams, but your data is even better, and thanks.

    Seattle Mike expresses my view pretty well. Even though the team’s record is a TEAM problem, and it doesn’t get better no matter who you blame.

    Perhaps individual and team pitching stats paint a clearer picture of the failures there, and batting stats are a little more opaque. So I’ve looked for ways of my own to assess how the Tigers are doing with the bat. I’m particularly concerned with how they hit with runners on (not just RISP). And the news isn’t good, with particularly surprising underperformance by Cabrera and Renteria. Even if failure to plate runners was roughly on par with league norms, we’re after wins, not league norms, and simply put, I think the offense has wet the bed more games than the pitching has.

  40. I would say the offense has more crapped the bed than wet the bed, Sean. Semantics, I know.

  41. Combining what I looked up with Lee’s figures, I think we can come to the conclusion the Tigers’ batters are doing their part, at least relative to the league, not expectations, despite everything. But the pitchers just aren’t getting the close wins.

    Here’s the record by runs allowed for low scoring games.

    0 runs — 0-0 (no shutouts!!)
    1 run — 1-0
    2 runs — 6-0
    3 runs — 0-1
    4 runs — 4-4
    5 runs — 2-7

  42. If anyone has the time, I’d like to see how the rest of the league fares in games where they allow either 4 or 5 runs. It seems as if a 6-11 record in those games is kind of bad.

  43. Lee

    Good info. Where did you get those numbers (so I won’t have to bother you for them again)? If you could, please put up the RUNS 5+/ALLOWED 5- numbers for all AL teams so we see how much of a correlation there is with the won-lost record. All I know so far is that that would make Boston 25-18 and Detroit 19-22. Pretty close, but surely there are anomalies, and looking at those teams might reveal something.

  44. Mark

    Oh, great. So I’m going to be sent to Toledo once some DTW poster currently on the “DL” gets back? Well, at least I’ll get regular starts. See y’all at Take I-75 North…

  45. Hey Kyle J,

    Sorry I didn’t give you a mention on the Spartan’s statistics – I’m a regular reader of your blog since I discovered it, which unfortunately wasn’t until tournament time. I read Big 10 Wonk religiously when John was still doing it, but you are doing an admirable job picking up the slack and I really like having a quality Spartan basketball blog.

    I got a chance to run the mean and standard deviation of all teams in the central, as well as the top and bottom teams in the other AL divisions (I used Angels as the top AL West team despite the fact that they are a half game behind the A’s). I ran them for 37 games. For teams that have played more than 37 games I chopped off the earliest games on their schedule. Also, my total runs scored figures for the White Sox did not quite add up to MLBs total runs scored, but I have double checked the data entry and can’t find any mistakes. Incidentally I am new at this so if these stats are available in tabular format somewhere could someone please point me in their direction?

    As I suspected, the Tigers have the highest average number of runs scored per game in the central at 4.757 runs per game, but also the highest standard deviation at 4.212. The corresponding figures for the rest of the AL Central are as follows:

    Twins 4.459 3.587
    White Sox 4.649 3.147
    Royals 3.595 2.522
    Indians 4.378 3.577

    For the other leaders/losers the stats are:
    Red Sox 5.27 3.106
    Angels 4.757 2.408
    Mariners 3.892 2.696
    Blue Jays 3.757 2.66

    No other team I ran stats on comes close to the Tigers standard deviation in runs scored – the closest teams round off to 3.6, while we are at 4.2. We are by far the least consistent offense in the sample – capable of going off but just as likely to tank.

    The Red Sox and Angels are notable for being both prolific offensively and consistent. For those two teams, even if they score a full standard deviation below their average, they will still score over 2 runs, whereas the Tigers would score around a half a run if they are one standard deviation below the mean. Couple that inconsistency with bad defense and pitching and you have a recipe for a pretty poor season. This is not an argument that offense is our main problem – our main problems are defensive – but we’d probably win a few more games if we could equal Boston’s or the Angel’s offensive consistency while maintaining our high run scoring average.

  46. OK based on Kurt’s link, teams not named the Detroit Tigers are 123-154 when allowing either 4 or 5 runs in a game (.444 winning pct.). The Tigers are 6-11 in those situations (.333). So based on that small sample of games you could say that the Tigs are fairly worse than the league average when allowing 4 or 5. So I guess that says something if you want it to.

  47. Chris, I have to agree that the offense has crapped and not just wet the bed. But I’m prepared to admit that I’m only half right and biased by expectations. Being half right would be progress for me…

    The Tigers record against RHP really jumps out at me. 11-21. But even that isn’t necessarily proof that the bats are the bigger letdown.

  48. Billfer should be ashamed for getting us all worked up over this. This kind of post is the equivalent of baseball porn for the stat geeks and Tiger junkies.

    We love it!

  49. Mark,
    Can you drop the high/low outliers for each team to see how much the 19-run game affects the Tigers’s standard deviation?

    And I agree this thread is like crack for us stat geeks. Must resist temptation to stay late at office and create Excel graphs on Tigers’ run production . . .

  50. The biggest problem is the offense, it is messing with the player’s and pitcher’s heads and causing everything else to tank. This was a team built on its offense, so when its main product isn’t going well it affects everything else.

    Its like if a pizza chain had horrible pizza, it would affect everything else, take Little Caesars for example.

  51. Billfer should be ashamed

    He has no shame. He’s a pusher. He’s probably hanging out by some playground right now, trying to get young kids started on OPS. Then it’s on to the hard stuff, WHIP and the like. I don’t even want to think about it.

    Back in my day, things were simpler. There was AVG/HR/RBI, and there was W-L/ERA. Kids face so many new pressures these days.

  52. Just wait until the kids start getting into VORP. That’s when it really hits the fan.

  53. I don’t even know what VORP is. Do I want to?

    If it helps prove that the Tigers hitting has been worse than their pitching, I’ll take it. But seriously… when I was trying to “massage” some of the game score numbers into showing that the bats were at the very least too inconsistent, I saw that you could say the same thing about the pitching. I’m going down in a swirling vortex of numbers.

    Two that stand out, as Ron mentioned earlier, are “16” and “22.”

  54. Sean C. in Illinois: VORP is Value Over Replacement Player and judges just that for each player. Obviously the higher score, the better. Here’s the wikipedia page for it.

    As for this new stat from Gage, it’s got nothing on a Mets blogger who blogs for the team’s website coming up with RBI/100 AB’s that somehow proved Brian Schneider was a good offensive catcher or something. Haha, good times. This is why I don’t read the media. That and their unwillingness (in Detroit) to hold people accountable for their blunders unless their name is Matt Millen in which it’s okay to publicly vilify him.

  55. Nice post Mark J on the standard deviation, and good stuff all the way around everyone else, nice thread.

  56. Haha Brian Schneider – Offensive Catcher. That’s akin to saying Pudge Rodriguez – Patient Hitter. On another note, I don’t mind seeing the words “Kansas City Royals” printed on the schedule for the next 3 games. They are a club who has seemed to come back to earth in recent times. Plus, if the Tigs can win the series they’ll claw out of the cellar. That would be nice.

  57. Oh there definitely needs to be some payback against the Royals. Nothing else has worked to get this team going, let’s try a little revenge beating, now that we’re coming up on another turn around AL Central teams.

  58. Mike R

    Thanks for the VORP def. Cool.

    Coming up on a Tiger game now, I guess, no time, but maybe tomorrow, in this thread I suppose, I would like to discuss a homemade metric of my own. not because I want to sell it as something for general use, but just to get comments on a) how it could be made better for what i use it for, and b) whether there is a metric already in use that does the exact same thing.

  59. Jeff M

    Followers of the Great Demon Lord Zoth sponsor his page. I’m suitably impressed.

    So, Truby is an example of replacement level.

    Reading up on the VORP concept elsewhere… well, I am astounded at the amount of thought put into things like this. Shouldn’t these people be curing cancer or discovering how to harness static electricity as an unlimited energy source? I shouldn’t talk, because I think about obscure baseball stuff myself. The difference is that I fail to come up with anything useful.

    But that’s only because my research here at the Institute For Totally Helping Everyone Now saps my strength.

  60. Hi Kyle J,

    I dropped the high scoring game from all teams in the sample (they all had a low of 0 so there aren’t any outliers in that direction). The Tigers still have the largest SD of any team in the sample at 3.51, and their mean runs scored drops to 4.36, putting us below the White Sox in average runs scored per game. The Twins and Indians (what is it about the AL Central?) have the next highest SDs at 3.33 and 3.14 respectively.

    Red Sox and Angels still look much better – their average number of runs scored only drops by .2 (ours drops by .4 from removing that one outlier) and their SDs tighten up a little more. Basically on a night where these teams perform a standard deviation below the mean they would still score more than 2 runs and we would round up to 1, but still be below 1. Seattle, Toronto, KC and the White Sox would all be over 1 run scored on an off night. The only teams below 1 on an off night other than us would be the Twins and Indians. The bottom line is that we still look pretty inconsistent (and have a low bottom offensively which showed up again in last nights game) even without that 19 run game.

  61. is running this (just a snippet):

    “The Tigers entered play Tuesday night as the only team in the AL Central with a lower batting average with runners in scoring position than its overall batting average.”

    Maybe Tom Gage is a contributor?

  62. Well it was nice to see Robertson get a little bit more luck this time out. It’s amazing what happens when you don’t walk anyone and keep the ball in the ballpark. Hopefully the other starters can follow suit with that line of thinking.



    SANTIAGO 37%
    INGE 22%
    ORDONEZ 19%
    CABRERA 19%
    POLANCO 18%
    THAMES 18%
    THOMAS 15%
    GUILLEN 15%
    RENTERIA 14%
    RABURN 13%
    JONES 12%
    JOYCE 10%



    JOYCE 20%
    INGE 15%
    ORDONEZ 14%
    RENTERIA 14%
    GUILLEN 14%
    THAMES 14%
    RABURN 13%
    JONES 12%
    CABRERA 8%
    THOMAS 7%
    POLANCO 7%



    INGE 17%
    THAMES 14%
    RABURN 13%
    CABRERA 12%
    JONES 12%
    ORDONEZ 12%
    THOMAS 11%
    POLANCO 11%
    GUILLEN 11%
    SANTIAGO 10%
    JOYCE 10%



    SANTIAGO 16%
    THOMAS 15%
    POLANCO 9%
    INGE 9%
    JONES 6%
    ORDONEZ 5%
    THAMES 4%
    GUILLEN 4%



    POLANCO 2%
    CABRERA 1%


    THAMES 31%
    JOYCE 30%
    THOMAS 19%
    POLANCO 18%
    GUILLEN 18%
    CABRERA 15&
    RENTERIA 13%
    INGE 11%
    ORDONEZ 8%
    RABURN 7%
    JONES 6%



    ORDONEZ 29%
    RABURN 27%
    POLANCO 26%
    RENTERIA 25%
    CABRERA 23%
    GUILLEN 22%
    SANTIAGO 21%
    JOYCE 20%
    THOMAS 19%
    THAMES 18%
    JONES 18%
    INGE 13%



    JONES 21%
    RABURN 13%
    CABRERA 12%
    THOMAS 11%
    INGE 11%
    RENTERIA 10%
    JOYCE 10%
    GUILLEN 8%
    ORDONEZ 6%
    POLANCO 4%
    THAMES 0%



    RABURN 13%
    JONES 9%
    CABRERA 8%
    GUILLEN 7%
    ORDONEZ 5%
    POLANCO 4%
    INGE 2%
    THAMES 0%
    JOYCE 0%
    THOMAS 0%

Comments are closed.