Guess the most potent team/game batting stat — ANSWERED

“One of these days in your travels a guy is going to come up to you and show you a nice brand-new deck of cards on which the seal is not yet broken, and this guy is going to offer to bet you that he can make the jack of spades jump out of the deck and squirt cider in you ear. But son, do not bet this man, for sure as you stand there, you are going to wind up with an earful of cider.”

Damon Runyon

That quote was in a little book I had as a teenager (and still have), called Never Give A Sucker An Even Break. Now, this challenge isn’t a sucker bet at all — in fact, I suspect most of you smart cookies will nail it — but I just couldn’t wait any longer to drop that line. Anyway, here’s the challenge:

  • Choose one team offensive counting stat that appears in a standard box score (excluding Runs and RBI).
  • The goal is to pick the stat which best predicts the winner in all 2013 MLB contests through Thursday 5/30, when one team had more of that stat than its opponent. When the team with more of the stat wins, that’s a win for the stat.
  • All games in which the foes had the same total in the chosen stat are kicked out of the pool.

To avoid any doubt which box scores are standard, we’ll limit the choices to:

  1. Plate Appearances
  2. At-Bats
  3. Hits
  4. Doubles
  5. Triples
  6. Home Runs
  7. Walks (including intentional)
  8. Strikeouts
  9. Hit By Pitch
  10. Sacrifice Hits
  11. Sacrifice Flies
  12. Stolen Bases
  13. Grounded into Double Plays

I have the 2013 winning percentages based on each of these stats, through May 30, via the Play Index and some Excel legwork. There’s a clear leader among these stats, by a margin equivalent to 5 wins per 162 games.

I’m not implying any meaning to the results. It’s just a game. I’ll show the percentages Tuesday, or whenever a sufficient number of suckers — er, deep thinkers — have made a guess. And let me know if you’d like the W% for any basic stat not listed here; if it’s part of a standard Team Game search in both batting and pitching, I can whip it up pretty quickly.

____________________

AND THE ANSWER IS…

  • “It’s gotta be hits.”

Below are the winning percentages for 2013 MLB games, through May 30, when a team had more of a given stat than its opponent. As a matter of interest, I’ve included in this table 3 stats that were not available choices in our game, which are shown in brackets: [Intentional walks], [Hits minus HRs], and [Unintentional walks].

[table id=127 /]

 

I was surprised that [Hits minus HRs] is so potent.

I did this little project because of a habit I started a couple of years ago. On the heels of some debate on the value of walks, I made it a part of my box score routine to tally in my head the daily record of the teams that draw more walks than their opponents. I never kept a running total, but I noticed that they almost never had a losing day. It took a while to figure out how I could assemble this information from the Play Index for a larger period of time.

Thanks to everyone for participating, and don’t let this be the end of the discussion!

(And P.S. to koma: What method did you use?)

0 0 votes
Article Rating
Subscribe
Notify of
guest

69 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
wx
wx
11 years ago

Stolen Bases. It’s no fun to pick something that everyone else is going to pick 😛
That being said, I did have Louisville in the NCAA basketball tournament so…

Doug
Doug
11 years ago

I have to believe it’s home runs.

But, seems like the obvious answer. Nevertheless, I’ll stick with the long ball.

RJ
RJ
11 years ago
Reply to  Doug

As you say, obviously the team with more home runs wins more games. But spurred on by the knowledge that the Giants were out-homered by their opponents last year yet still won the World Series, I thought I’d see if any other teams had success sans sluggers.

In fact, San Francisco was the only team last year to be both out-homered and post a winning record. Philadelphia had a HR defecit and finished at .500 on the money. Pittsburgh and Miami outslugged their opponents but had losing records.

bstar
11 years ago
Reply to  Doug

Kinda difficult for a Braves fan to not jump on the HR bandwagon here.

Atlanta’s W-L record when they homer/don’t homer is scary. It’s 2012 Yankees scary. Maybe I’m mistaken, but I believe Atlanta’s only won 3 games this year going homerless.

Yeah, I’m sticking with HR.

Jimbo
Jimbo
11 years ago

I’ll take walks.

Jimbo
Jimbo
11 years ago
Reply to  Jimbo

I as going to take plate appearance, but home teams don’t bat in the bottom of the 9th when they win.

Walks tend to indicate either the pitcher lacking command, or the batters having discipline, or a combo of the two.

RJ
RJ
11 years ago

When you say “choose one team offensive counting stat… that best predicts the winner”, do you mean the team with the most of that stat has to be the winner? For example, I reckon teams who ground into more double plays than their opponents lose more often. Therefore GIDP would be predicting the winner… it would just be predicting the OTHER team as the winner. Get me?

Otherwise this sucker is going to say Plate Appearances. Seems like having more guys take a turn at batting would be a good way to win some ballgames.

RJ
RJ
11 years ago
Reply to  John Autin

I didn’t think you meant it like how I suggested, but given your opening anecdote I’m on extra lookout for shenanigans. 🙂

Phil
11 years ago

Plate appearances, same rationale as previous post. Wild guess of doubles at #2—makes me think of a good-contact, hard-hitting team, not prone to slumps like a one-dimensional HR team (I’m from Toronto…).

ReliefMan
ReliefMan
11 years ago

Not that it’s very helpful here, but teams who receive more RBI intentional walks than their opponents in a game have an all-time W% of .000. Maybe managers will start catching on to the tactic and letting their pitchers use it a lot more often to ice a win.

no statistician but
no statistician but
11 years ago

Given that this isn’t a trick question (as the Damon Runyan quotation suggests it might be) along the lines RJ indicates or some other, I’ll go with plate appearances, since, in my own non-Play-Index, non Excel, primitive way, I’ve looked at the standings and the basic team stats, and the teams with winning records tend to have more plate appearances than the losers, even though most of their home wins come with only 24 outs. This is not as simple minded as it may seem, since, while the number of outs in a game is a constant—all right, some go… Read more »

Luis Gomez
Luis Gomez
11 years ago

I will go with plate appearances. I pretty much think among the same lines as NSB, but I´m too lazy to explain myself.

Luis Gomez
Luis Gomez
11 years ago
Reply to  John Autin

Thanks for making me spit my coffee all over my keyboard. That quote at the end by Mantle made me laugh out loud!

(But I´m still too lazy to read Casey´s entire testimony!)

Hank G.
Hank G.
11 years ago

I would have guessed LOB, but you didn’t include it as one of the possibilities. Isn’t LOB usually in the box score? My impression from my youth was that despite LOB supposedly being a bad thing, the team with more LOB was usually the winner. I have no idea whether that is actually true.

e pluribus munu
e pluribus munu
11 years ago

I’m with RJ and Phil: PAs, despite Jimbo’s caveat.

By the way, Mr. Runyon’s pearls were lifted from his idyll by some lightfingered characters and dropped word for word into the book for a Broadway song and dance that you can find trotted out in many a metropolis to this day.

koma
koma
11 years ago

all games sorted by each of these stats and then taken the top 300 results shows, that the most games are won with a lot of hits, so i think it were the hits.

Hartvig
Hartvig
11 years ago

I’m going to go with koma on this and say hits as well. I’m pretty sure it will turn out to be one of the 4 already mentioned- excluding wx’s contrarian view- PA’s, hits, homers or walks. I was going to try and explain exactly why I came to that conclusion but since my thoughts have been all over the place on this one but I’m not sure it would make a lot of sense- suffice it to say that run scoring environment and how we calculate OPS were the deciding factors in my decision. Now I can go to… Read more »

Bryan O'Connor
Editor
11 years ago

I’ll join the masses and take plate appearances, about 11 spots ahead of sacrifice hits.

brp
brp
11 years ago
Reply to  Bryan O'Connor

My first thought was PAs, but if you’re winning at home you’re not batting in the bottom of the ninth (as Jimbo said at 5). Not sure if that makes enough difference to counteract the fact that you’re scoring more runs than your opponent, but…

I’m still going to stick with PAs. I imagine walks have a good correlation as well but walks count toward PAs, too.

Mo
Mo
11 years ago

Plate Appearances

Abbott
Abbott
11 years ago

I’m going to go with Hits. It’s gotta be hits.

Kenny
Kenny
11 years ago

Tongue in cheek, I like grounded into double plays…at least it suggests that the winning team was able to get runners on base. Ha.

GrandyMan
GrandyMan
11 years ago

I’m going to jump on the Walks horse.

Rocco
Rocco
11 years ago

I would say that if a team out-homers another team, it has about a 75% chance of winning the game. So home runs it is.

BryanM
BryanM
11 years ago

Gimme triples. Rare, but almost always well hit balls

Brendan Bingham
Brendan Bingham
11 years ago

Doubles. It’s highly correlated with runs scored (better predictor of runs scored than HR, I believe). Whether it’s better than hits or PA, I don’t know. Nonetheless, I’m saying doubles.

bstar
11 years ago

Brendan, that’s really interesting. Is there a logic behind it? Is it because more doubles are hit than HRs?

Timmy Pea
Timmy Pea
11 years ago

Hits include HRs, triples, and doubles. If you get a HR you are also credited with a hit. I say hits are the most important. What’s better, a team that has 3 homers and 6 total hits, or a team that 3 homers and 9 total hits? Of course it’s the latter. OK, I know the argument will be what about 1 homer and 9 hits vs. 2 homers and 6 hits? Well since we’re not allowed total bases as a choice, I still stick with hits.

Hank G.
Hank G.
11 years ago
Reply to  Timmy Pea

Plate appearances would incorporate walks and hits, so I think that’s got to be better.

Timmy Pea
Timmy Pea
11 years ago
Reply to  Hank G.

PA’s also have strike outs and fly outs.

RJ
RJ
11 years ago
Reply to  Hank G.

I am starting to think that those missing plate appearances in home wins (as noted by Jimbo and others) matter quite a bit. That’s a large chunk of games where the winning team has an automatic defecit of 3 PAs. In home wins by a large margin, that won’t matter. But those will also be games where the home team has likely outhit and/or outwalked the opponent. In close home wins though, lets say where the home team has only outhit the opponent by one or two hits, they may not have enough PAs to overcome not batting in the… Read more »

koma
koma
11 years ago

hits are a 80,4% correct predictor.
3 points better than HR.
And PA come in 3rd with 69,3%

Timmy Pea
Timmy Pea
11 years ago
Reply to  koma

Thanks Koma, what do I win?

Timmy Pea
Timmy Pea
11 years ago
Reply to  John Autin

Well a hit is almost always earned and it’s never guaranteed. You are guaranteed at least 27 PAs per team every game. That waters down it’s value. PAs are important, but there are just too many low scoring games in which they really aren’t a factor to put them in front of hits or homers.

Kenny
Kenny
11 years ago
Reply to  John Autin

Gee…never been called a hanging chad before…

no statistician but
no statistician but
11 years ago

JA: What’s really unexpected in your figures—and why so many of us were wrong to choose plate appearances—is the relative low impact of walks in the results, or low compared to hits. But it makes sense, and it supports several notions: 1) a walk, while good, isn’t as good as a hit, so the TTO theory is, as some of us keep saying, a mistaken view; 2) a hit has added value because of its potential for more than a one base advance by the hitter or by any men on base at the time; 3) a walk can result… Read more »

Timmy Pea
Timmy Pea
11 years ago
Reply to  John Autin

Stop comparing a walk to a hit. Start comparing a walk to an out! A hit is a little better than a walk, but a walk is about 3 times better than a strike out.

brp
brp
11 years ago
Reply to  Timmy Pea

Way back in the late 1980s I read a Stats Inc. book that broke things down and said “a walk is 2/3rds as good as a hit.” This certainly seems to bear that out.

Timmy Pea
Timmy Pea
11 years ago
Reply to  brp

I should clarify, I’m comparing a walk to a single.

no statistician but
no statistician but
11 years ago
Reply to  John Autin

JA: Grumble grumble. Earlier I sent a short response to your disagreements which failed to make it to HHS and so now is winging it in the general direction of Orion’s Belt, I suppose. 2nd try: Re: TTO, “theory” may be the wrong term. How about “justification,” as in the justification for the hitting approach of Adam Dunn and his ilk. Re: point 4, I wasn’t thinking of IBBs—your chart above makes it clear that that horse won’t run—but of semi-intentional walks to certain dangerous batters—not giving a Ted Williams anything good to swing at. let’s say, after the count… Read more »

Jimbo
Jimbo
11 years ago

I guess the 2001-2007 Giants really skew the data. Because they never went anything like 131-31. So I’m guessing if you remove them, IBB’s become a much bigger predictor.

RJ
RJ
11 years ago
Reply to  Jimbo

Jimbo, the data given is only for MLB games in 2013 through May 30.

Chuck
Chuck
11 years ago

The “Walk Fetish” World Tour has been cancelled effective immediately, with no re-scheduling likely in the future.

Thank goodness, because I’ve seen it a few times and it sucked.

Artie Z.
Artie Z.
11 years ago

JA – can you run these numbers for singles? I’d like to see where the team that has the most singles ends up in that list.

koma
koma
11 years ago
Reply to  Artie Z.

singles were correct 67%

koma
koma
11 years ago
Reply to  John Autin

sorry for the misplaced reply, but i normally forget to refresh my browser before posting my comment, so i first missed your changed post on top with the solution and then missed your comment asking for my method;-(

koma
koma
11 years ago

@JA

i cheated a little
i think i did the same as you
team batting game finder “home games”
team pitching game finder “home games”
sort ’em both by date and home team(in open office)
and look after the difference of each stat compared to the result.

once the copy/paste thing was done(which was indeed the most part of my work, because of the 300 lines limit), the rest was easy.

birtelcom
Editor
11 years ago

A major shortcut that doesn’t do exactly what John is doing but would likely produce, in bulk, similar results, would be the following: We know that the MLB average number of walks per nine innings is a bit over three. It’s been over 3 every season since 1968, but is rarely over 3.5. This season so far it’s been just slightly over 3. From that we can gather that, on average, a team that walks 4 or more times in a particular non-extra-inning game will out-walk its opponent (not always, of course, but on average), whereas a team with 3… Read more »

Timmy Pea
Timmy Pea
11 years ago

John this was a great post! This is not meant as a dig, but in my mind I picture a guy that knows every rule and regulation of the United States tax code. I don’t mean it in a bad way, it’s just the way you process your information and pass it along.

Jimbo
Jimbo
11 years ago

how about most total bases?

and something odd, like most balls put into the air, or on the ground?