20-13, in first place by 1.5 games.
Is this the first Saturday night game of the year? Seems like it.
Phil Coke has been activated, and Luke Putkonen has been optioned back to Toledo. I was thinking Alburquerque was next in line for Toledo, but instead the Tigers may opt for the ever-popular retroactively injured thing next time he struggles with his control.
*********
Some of my favorite “newfangled” baseball stats are WPA (win probability added) and RE24 (base-out runs or runs saved). They’re all over the place here, and I especially like to look at them in player game logs, the box scores, and the full play-by-plays (scrolling down in the box scores). If you don’t know what they are and mean, well, check it out. You’ll figure it out.
I’m not much for WAR. To me, it’s a fantasy baseball stat, though I’m aware that GMs and agents playing the real-life version of fantasy baseball actually use it. The thing I love about WPA is that it allows you to look at contributions to wins/losses on a PA/BF by PA/BF (plate appearance/batter faced) or game to game basis. It gives you about the most objective “game score” for a player that I can imagine. RE24 is a kind of productivity score along these same lines, though not one tied to win/loss probability based on before/after game situation. WPA and RE24 can be looked at in the aggregate like any other stat, of course, but then you have to consider adjustments for opportunity, same as you would when comparing HR or RBI totals between batters with different numbers of PA. I find them useful in any event, but most interesting as ways to gauge performance in individual games or plays within games. The only drawback I find is that they cannot really account for many significant positive defensive plays, though I suppose you could invent some tweaks to introduce defense as more of a factor. (The difficulty is mainly that it’s hard to quantify “what if?” situations.) So it’s mostly a batter/pitcher kinda thing, the defense being an assumption.
I keep meaning to post some interesting analyses that utilize WPA and RE24, like this one for 2013 Tigers hitters through 32 games, for instance:
WIN CONTRIBUTION PER PLAY (SCORE)
CABRERA +955
FIELDER +512
TUIASOSOPO +487
HUNTER +435
JACKSON +185
KELLY 0
PERALTA -073
INFANTE -167
AVILA -297
DIRKS -306
MARTINEZ -606
SANTIAGO -1481
PENA -1500
PRODUCTIVITY PER PLAY (SCORE)
TUIASOSOPO +188
CABRERA +138
FIELDER +102
HUNTER +65
JACKSON +19
PERALTA +12
INFANTE -13
DIRKS -16
KELLY -58
MARTINEZ -67
AVILA -95
PENA -130
SANTIAGO -178
But each time I try, I collapse under the sheer weight of it all. I doubt that I’ll ever have time in between games to work up the kind of comprehensive yet approachable overview I aspire to. The fun thing about it is that hidden truths are revealed, some of them contrary to what the more visible aggregate stats are telling you. Such as, to simplify greatly: Don Kelly isn’t really “crap,” comparatively speaking (I’m looking at you, VMart), and Doug Fister hasn’t been as good as you might think. Also, looking back on a box score and scrolling down to the “Top 5 Plays,” you may discover that “oh yeah, that was kinda cool” was actually the most significant play of the game in terms of WPA. All in all, it just gives you a broader view of where the positive and negative contributions are coming from. The cool thing about RE24 in a box score is that one number summarizes the productive sum total of every PA or BF. All “1 IP, 0 ER” and “0 for 3″ and”1 for 4” are not created equally.
So check it out.
*********
Tonight’s “Victor Martinez is available to pinch hit for Victor Martinez” lineups:
DETROIT – It’s the A-team again, behind JV, no less
CLEVELAND – Raburn-less again, go figure
CF Bourn
2B Kipnis
SS A. Cabrera
RF Nick Swisher
C Carlos Santana
DH Giambi
1B Reynolds
LF Brantley
3B Chisenhall
*********
It appears from the latest forecast, which I seldom interpret correctly, that there may be a rain delay and/or rain early in tonight’s game.
POST-GAME: INDIANS 7, TIGERS 6. Cleveland was hitting Verlander’s fastball at will from the get-go, and Jimenez was just too good. You could see this one going down the tubes on 2 plays: a) Tigers 4th, Miggy and Fielder on 2nd and 3rd with 1 out… Martinez grounds out to 1B, no chance for a score, and then Dirks strikes out way too easily; b) Indians 5th, what starts as a great play by Fielder to step on 1B and then double up Swisher in a rundown ends with Swisher (who soon scores) at 2B on a ridiculous throwing error by Miggy. Then Smyly and Albuquerque crap out, score going from 4-1 to 6-1 Indians. All over, right? NO! VMart starts a sweet 4-run rally that includes a couple good PH at-bats from Tuiasosopo and Pena. Then… the sickening Al-Al lets in an 8th inning Indians insurance run before Downs saves the day with bases loaded. Tigers 8th, and things are cooking until Tuiasosopo grounds into a 2-on, inning-ending DP. With Perez in for the Tribe in the 9th, it’s not looking good until a Swisher error puts Pena on 1B with one out. Hunter delivers a 2-out RBI to make it 7-6 Indians, and suddenly the win seems not only possible, but likely. Because MIGGY is at the plate with men on 1st and 2nd. A truly lousy swing and miss on pitch 2 turns this into a crap AB, and his weak grounder to the left side overtakes the Tuiasosopo GIDP as the big plate failure of the game (Tuiasosopo actually had a pretty decent AB overall there). No joy in Mudville. Very exciting game, though. Kipnis’s incredible play for the force at 2B and the second out in the 9th on the Jackson almost-single up the middle was the game-saver for the Indians.
PLAYER OF THE GAME: Ubaldo Jimenez, Jason Kipnis
HONORABLE MENTION: Darin Downs, A. Cabrera, Michael Bourn, Victor Martinez, Omar Infante
NOT SO GOOD: Al Alburquerque, Indians bullpen, Justin Verlander, Miguel Cabrera, Andy Dirks
I knew Putkonen was doing too good!…..see ya!…..Mumbles was quoted as saying “We need Coke in that LH Specialist role”……….oh! I didn’t know you ever used him that way!!!!!
Good point. We’ll see if Coke becomes that lefty specialist now. It’s not as though they can’t afford to bring him in for one or two LHB and then pull him, with two other lefties in the pen, both well established as 2 or 2+ innings guys.
I agree with you that WAR is more of a fantasy statistic. Leave that for the managers to decide how they want to interpret it and what to do with it, but as a fan you don’t really care about what a player’s WAR number is and you don’t identify with him as a specific WAR level.
These stats are fun but you have to be careful with any statistic that tries to tell you about Win Probability as that opens the door to a lot of assumptions and opinions based upon what input parameters you use to determine how the odds are calculated. I have for years been checking the AdvancedNFLstats.com site which is pretty well known, and they have a lot of probability machines for season predictions as well as a live game chart that can show you how Win Probability changes with the result of each play. But after years of keeping an eye on it I have noticed lots of funky things with how their probability machines work and the numbers can be drastically different if you alter the perspective, so you have to take this data with a grain of salt. At the end of the day you can’t compute a Win Probability without making assumptions as to what a given player or situation is worth, and from my experience baseball stat geeks often have disagreements about what those values are. The numbers for one already tell you only an arbitrary and somewhat abstract story, and then the numbers are only as good as the formula that is programmed to compute them. Combining those challenges can make it tricky to identify with what is this data really saying and what do you do with it anyhow.
But one things for sure, it is fun data, and it’s fun to see Tui clobbering Cabby, that’s a good wookie!
OK, I am at peace with this, but this is now two days in a row where I find myself agreeing with my young fellow blogger TSE! Is this some sort of streak? And if so, do I have to wear the same shirt, or socks or what? Seriously, when I took a Stats class in College (in the 70’s!) I do remember the prof saying that if you try to compare apples to oranges it won’t compute. Or as Charlie Borown once said, Tell your stats to shut up!
Ha! Ok, well for that then we’ll tell the WAR stats to shut up and I’ll put up my stats! Here is my OE% of the Tigers in 2013 thus far to match up with SL’s post today so we can see a similar stat to the one he has, which is the Productivity Per Play Score.
My OE% is similar to that. It is a measure of how efficient a player performs when in the batter’s box. Notice how I don’t say how great the hitter is. It’s a blended catch-all stat that seeks to find who has been efficiently contributing and the best players at performing or making good decisions will help increase the score. The stat isn’t visually interpretive like a Slugging score is. If you Slug .500 we can imagine a base value for what that means on average, or a .300 BA we can visualize the frequency that a player gets a hit, but this is more of a composite, kind of like OPS whereas it’s a score. A good range of scores will generally exist in an area of numbers that’s between a good BA and a good SLG. LIke an OPS, a player is never going to perform in a given plate appearance for the specific OPS or OE% that he averages as there’s not one specific result that is ever going to equate to a player’s unique 3 digit average score per plate appearance.
Generally I look at .450 and up as “good”, .400-.450 as “adequate”, and below .400 as “failure”.
So here are the numbers so far through 32 games:
Cabrera .618
Tuiasosopo .595
Fielder .565
Hunter .447
Infante .440
Peralta .423
Dirks .402
Jackson .382
Pena .312
Avila .300
Kelly .293
Martinez .290
Santiago .239
TSE, I’m afraid I’m going to have to ask you to stop being reasonable. It’s throwing everything off here.
Seriously, good stuff, thanks for the replies. I’ll get back to you on this, no time at the moment.
Ok you’re the boss!
To me the goal is to find as many players as you can string in a row that can all perform efficiently in coordination with each other at a good percentage of the time. When you can create a chain link of positive events, you can get more powerful results than the sum of the face values. Like if it takes 4 bases to score one run, well if you only had 2 in one inning then you could have an infinite number of 2’s and never score a single run. But when you can combine 2’s in one inning, then that second one becomes more important only because of the first one that precipitated it, since that would result in a scored run. So if you can get enough of a consistency of highly efficient numbers, then you can not only get a bonus for having stronger input numbers for each individual plate appearance, but you can get extra “bonus round values” on top of it by connecting successful events with each other.
I just made that term up since people in Detroit seem to identify well with bonuses as I’m picturing all of those slot machines where there’s infinite different versions of bonus rounds so that should be a term people can really appreciate. You play the slot not for the single line score, but to hit the multipliers and line everything up so you can get into those jackpot scenarios. You want to brake the bank and you need that special scenario for the exponential bonuses. And of course you have to pay 3 chips, not one, if you want the extra bonus on top of the normal bonus. And creating a solid 1-9 lineup of efficient hitters is like juicing the slot machine to put yourself in a position to maybe collect on those best possible winning scenarios.
What if I told you I had a weigh you could get a fourth line that isn’t normally printed on the machine? You would salivate at the math of having an extra line on top of the normal lines. And so creating that solid 1-9 lineup is in a sense a way of rigging the machine so that it pays out too much, unlike a regular slot machine, aka a normal baseball team, aka the way everybody else does things.
And the OE% requires a complete effort as it’s an overall based statistic. So if you are a high average guy with no power like Torii Hunter, then you have one variable you might be weaker at than another. A guy like Ryan Raburn who has average and power right now might lose a couple points for still striking out a lot as that is maybe his weakest statistic.
So for a general game, a good lineup of 9 strong overall OE% numbers is what you want. Then in specific situations you can break the stat down to have specific variables used as modifiers. For example if you were in a situation where the bases are loaded and you wanted to rank available hitters by OE%, well you could tweak the OBP variable of that score and make it more important and then feel comfortable than you not only have a high overall OE% performer, but a guy who ranks high when you make OBP have a higher weight of importance. If you wanted to just score a guy with a sac fly, well then you filter another way and you could make add a bonus modifier for sacrifice percentage to see who not only has a good OE% overall, but also one where a sacrifice talent is heavily weighted in favor for. Because you can score that run with a sacrifice, but you can still also score it with a single or a double or a homer or whatever else the OE% normally would account for. So you have different ways to mesh the data as a generic measure in contrast with any isolated other element.
way not weigh wow
TSE –
I think the idea of a chain link of positive events is way too complicated. Not to mention really out of a player’s or manager’s control. The idea of a batting order is of course an attempt to get a chain link of positive events going, but it can only do so much.
TSE –
Your OE% is interesting and possibly more original than my “productivity score” up top, which was nothing more than cumulative RE24 divided by number of “plays” (mostly PA with some baserunning events, I think).
Kinda wondering why Jackson and Kelly score so low on yours. and Pena and Dirks so high.
Well Kelly is low because his BA and SLG scores are pitiful. AJ also has pitiful power this year with only 3 hits better than a double, that’s only 1 triple and 2 HRs, and when you have a weakness in power you need a strong BA to make up for it and AJ has a paltry .280 AVG which is horrible when meshed with no power.
That’s part of the reason why I hated the Hunter signing so much. Look at how not impressive Torii’s score is compared to the big guns, and that’s with his .340 batting average. We didn’t know Torii would be batting .340 when we signed him, and it’s taking a monster BA for him to maintain a reasonably good OE% score. If he can’t eclipse that .450 mark with that BA, then how bad could it drop if he only bats .280 like AJ has been. That’s a scary thing to think about since we have a very small number of productive players.
Pena isn’t really that high, he’s pretty close to the bottom. Dirks is only about a .400 score, and if you compare him to Hunter you can see Dirks has a lot more power with 4 3B/HR compared to only 2 for Hunter. So he has double the amount of big power shots as Torii, and he has done that with only 86 atbats to Torii’s 133. So that helps Dirks a lot, if he had a BA like Torii then he would be off the charts, but he isn’t hitting frequently enough to get with in range of Torii’s score, but he has enough potency of power to at least get to that .400 level.
TSE –
WPA isn’t a magical stat that tells the whole story. No stat does. What’s to like is that a) it’s about winning a game vs. losing a game, nothing more and nothing less, and b) it doesn’t care about WHO, but only about WHAT and WHEN. It also doesn’t care about WHAT IF, which is kind of a mixed bag. There’s nothing tricky or funky about the way it works, though.
Based on the play-by-play results of 10 gazillion MLB games since the dawn of time or thereabouts, the team win probability for any given inning/outs/baserunners situation can be calculated with a high degree of accuracy. Every play will change this situation, and by comparing the before and after, the effect on win probability is quite accurately determined. The disagreements of stat geeks do not enter into this, as far as I can tell. It is empirical fact and not mere theory that if your team is batting in the bottom of the 9th, down 7-6*, 2 outs, men on 1st and 2nd, you have a ~17% chance of winning the game. Doesn’t matter who’s playing, who’s pitching, or who’s batting. These factors are as spitting into the ocean. As these particular situations pile up, the odds might start to diverge from 17%/83% or whatever. But very slowly. And it will, or course, be known by the time they get to 18%/82% or whatever.
What WPA can’t account for (besides all the things for which you must watch the game to know) is the effect of defense, unless it involves an error. In the case of a leaping catch that robs a hitter of a home run, well, if we had the “WPA machine” that spits out the odds, we could make our own calculations for the WPA value of that run/runs-saving play. Other defensive plays would be less clear-cut. Some great plays clearly prevent some sort of 1-base hit, but many are trickier. You would have to bring assumptions about alternative results into your WPA calculations. Which is not to say that it couldn’t be done. But it wouldn’t be as strictly objective as the pure WPA as we know it.
* I think actual score is involved and not just run differential, but I could be wrong.
Actually, my explanation of WPA isn’t quite right. I think I reasoned my way to an understanding more than I properly read my way to it when I encountered WPA several years ago, so here’s the proper explanation:
http://www.baseball-reference.com/about/wpa.shtml
Hey Mumbles…Avila has had a couple good games in a row…shouldn’t he be sitting today?
As Sparky Anderson used to say…can make those stats tell you everything I want you to know!
Nick F. Swisher, new Tiger Nemesis. (And Ohio State guy btw).
I’d rather get beat by a guy like Swisher, a fun guy, than Denard F. Span. I don’t like Span’s demeanor. It’s insolent and unsettling, and I think Tigers pitchers may find it upsetting or at the least distracting, which is unacceptable. I’m going to write to the commissioner about it.
hey JV feel free to get some of these guys out…….WTH??????
Holy Dontrelle Willis, wake up the bullpen!
Ok, who is the joker who told Verlander this was the All-Star game?
Notice lately as Jacksons average keeps slowly slipping downhill he is back to swinging at the balls and taking the strikes!
First base is open, time for the one-pitch walk?
This is going to be one of those pray-for-rain games, isn’t it?
I suppose we were due for an outing like this from JV….been pretty rare!
and as much as Jimenez has been struggling he will turn in to CY tonight to add to our misery
Would be nice if he could get a breather, but apparently we are up against Cy Jimenez tonight.
Actually Verlander hasn’t been particularly efficient yet this year (note the high WHIP). He’s been throwing a lot of pitches all along. It just caught up to him tonight.
Yeah, pitch counts getting high too early all or most of the time. I’ve noticed.
Funny how they always talk about Jimenez walking all those batters….but the Tigers make sure that isn’t going to happen…….at least 7 of the 12 outs are on balls so far
hey Victor why don’t you just throw your bat at the ball…..oh excuse me you DID!
You can’t WIn them all….but this is some darn ugly baseball on the Tigers part…
Rough night. I’m at the game – can’t win them all, but not much energy here tonight.
I confess I was hoping for rain after the first, but I didn’t know you were there.
I blame Leyland. He slipped up and ran out the A lineup while Verlander was pitching.
Oh it rained. And it’s cold.
Not JVs night. He’ll rattle off 8 QSs in a row or something like that here soon.
Long season.
The thing I hate most about this game is|
Before the game The Tigers knew they couldn’t go out of the strike zone and be successful against Jimenez
During the game they knew you can’t go out of the strike zone batting against Jimenez
After the game they will say …..we should have shown more patience and not swung at so many pitches out of the strike zone against Jimenez
hello!!!!
Doctor! The patient has a pulse!
Ladies and gentlemen, I present to you the best #9 hitter in Major League Bsseball.
In both leagues I may add…and he is coming to bat in the 9th!
I hope mcb is still there.
too little too late….still an ugly game….final answer. Played two out of 9 innings
They managed to work themselves into a tying run on 2nd Cabrera at bat situation…not bad. Gotta like the way Omar is handling the bat. And Torii “The Professional” Hunter (Jackson, not so much).
Is first base really that easy to play? Someone ask Nick Swisher.
Stranding Peralta earlier in the game, especially at 3B, came back to bite.
I was afraid we might be in trouble when Kelly, Santiago and Pena weren’t in the starting lineup.
Oh I stayed!…nine hard innings!
I’m pretty pleased to uncover this site. If you don’t mind May I also share a word of advice. Fidning a Personal sexy muscles love firm could well be dificult, in case you reside in Sourthern California therefore you need a Personal sexy muscles nude man at love, click on my link.