THE BOOK--Playing The Percentages In Baseball

Wednesday, March 18, 2009

Offensive rates as a function of base/out state…

I have spend a lot of computing time (it takes forever to “run") trying to figure out what the league average rates for the various offensive categories are as a function of the base/out state.  Any differences from overall rates would be due to batters and pitchers changing their approach and the fielders being in different positions and perhaps doing different things when the ball is hit to them.

It appears from the data that the batters trump the pitcher in terms of their approach (although, as you will see, there are times when the pitchers truno the batters, as with BB and HP rates), or perhaps the pitchers don’t “care” as much as the batters, and perhaps rightfully so.  For example, with a runner on second base and no outs, batters in general (lefties and righties, BTW), do in fact hit more balls to the right side, even though to some extent the pitcher is trying to “force” then to hit to the left side.  We mostly see that effect with no outs, BTW, which is pretty much what we would expect.  If a commentator ever tries to say that batters try and hit to the right side with a runner on second and ONE out, don’t believe him (it is certainly possible that that happens SOME of the time - but on average, there is not much evidence of that).

For all situations other than at least a runner on second and no outs, batters hit fly balls to the left side 51 or 52% of the time and ground balls .55 or .56.  This is RHB and LHB combined.  With no outs, however, and a runner on second only, it is .47 and .48 (FB and GB), respectively.  With first and second and no outs, it is .53 and .52, so fly balls are normal and ground balls are hit more to the right side, I guess to help stay out of the DP, or maybe as a function of the hit and run.  With 2nd and 3rd, it is .5 and .55, so ground balls are normal and fly balls are a little to the right side.  With a runner on third in general, we see more fly balls to the right side.  My guess is that with zero and one out and a runner on 3rd, the batter is trying to hit a fly ball, which results in more fly balls to the opp field for both lefties and righties.  In fact, with a runner on 3rd only, we see .48 for FB.  With one out, we see .5 and with 2 outs, we see .51.

Anyway, I was most interested in the rates of s,d,t,hr,bb, and k, given the various base/out states.  I don’t know how I can show you the entire results without linking to a spreadsheet, but here are some highlights:

Everything is presented as a ratio of what happens at a certain base/out state compared to overall (all base/out states combined).

With no one on and no one out, we see more extra base hits, including the HR and more triples, even per non-HR extra base hit.  We see a lot less BB and HP.  (BTW, we see the HP mimic the BB in terms of the base/out state, so it appears that the HP is not nearly a random event!) We also see slightly less SO.  Here are the numbers:

s 0.99
ex 1.03
d/ex 1.00
t/ex 1.04
hr 1.07
bb 0.89
hp 0.89
so 0.96
roe 0.91

BTW, you see many fewer ROE with no runners on base than with runners on base, for all numbers of outs, for obvious reasons (more pressure on the fielders.

But wait, you might ask, what about the batters’ overall talent for this base/out state? Isn’t the pool of batters overrepresented by lead-off men?  Sure.  So I adjusted for the pool of batters faced at that base/out state.  (My next step is to adjust for the pool of pitchers for each base/out state.)

Here are all the batters that hit in that state, as compared to all batters:

s 1.01
ex 1.00
d/ex 0.99
t/ex 1.09
hr 0.97
bb 1
hp 1.01
so 1
roe 1.01

As you can see, a pretty normal group of batters, other than HR rate.  So the “effective” HR rate for this base/out state is not 1.07, but 1.09! You might also say, “This is a little circular.  The batters in this pool will tend to see more of this state than other states, so their overall stats are affected as well.” And you would be right.  I don’t think it makes enough difference to want to do a recursive analysis, but you could to be more accurate.  And of course the fact that batters tend to change their batting order position tends to mitigate this “circular” effect.

What about an “extreme” base/out state like loaded and 1 out. What do batters and pitchers do?  It is important to keep in mind that batters and pitchers don’t HAVE to change their approach.  They do so voluntarily, on order to leverage the base/out state, at least presumably so.  I guess it is also possible that there are things that they do subconsciously with various base/out states that are not necessarily optimal.  And, as I already mentioned, part of the change in rates (s,d,t, etc.) is due to fielders being in different positions.

Here is what we get before adjusting for the pool of hitters:

foulout 1.11
s 1.02
ex 1.03
d/ex 1.01
t/ex 0.86
hr 0.91
bb 0.8
hp 1.38
so 0.94
roe 1.70

Lots of non-average things going on, as you can see.  Batters and pitchers do not like walks.  The HP is high though. I suspect that is because some batters will let themselves get hit with the bases loaded, since they know it is a run and an RBI.  Very high ROE.  I assume that is because there is a lot of pressure on the defense and because a small percentage of this base/out state is with the IF in.  That is also probably the reason for the higher singles and doubles rates. Low HR rate.  That is probably due to a combination of the batters trying to make contact and not striking out, and pitchers being loathe to give up a granny.  I am not sure why so many foul outs.  Maybe batters desiring to make contact and not strike out more. Let’s make sure there is not an unusual pool of hitters driving these unusual numbers.

foulouts 1.01
s 0.99
ex 1.00
d/ex 1.01
t/ex 0.94
hr 1.02
bb 1.00
hp 1.01
so 1.02
roe 0.99

We seem to have a pool of slight power-type hitters here - high HR and K, low triples.  So the low triples rate in this base/out state is not really that low after adjusting for the batter pool. The low HR rate ends up even lower and so does the low SO rate.  Here are the adjusted rates:

foul 1.1
s 1.02
ex 1.03
d/ex 1.01
t/ex 0.92
hr 0.9
bb 0.8
hp 1.36
so 0.92
roe 1.71

OK, one more:

Runner on second and no outs:

Since the batters (on the average) are obviously trying to move the runner along and in doing so have to alter their normal swing, let’s see what happens overall. Unfortunately, I did not remove pitcher hitting, so it includes pitchers bunting.

foulouts 0.94
s 1.05
ex 0.98
d/ex 1.00
t/ex 1.02
hr 0.82
bb 1.14
hp 1.38
so 0.86
roe 1.21

Looks like in an attempt to move runners over in general, batters get more singles, but much fewer extra base hits and HR.  Pitchers, as you would expect, walk a lot of batters (BTW, these walks do NOT include IBB) and hit a lot of batters.  In this case, with the bb and hp, the pitchers trump the batters as the batters are not trying to walk necessarily (although it is never a bad thing to walk with no one out, no matter what the base situation - in fact, I suspect that pitchers are walking too many batters in an attempt to be too “fine” with a runner on second and a base open).  Interestingly, there are few K even though pitchers are probably trying to K the batter and are also throwing a lot of pitches outside of the K zone (hence more walks).  The batter trumps there I guess since he is really trying hard not to K.  And as usual with runners on base, there are many more ROE’s.  For some reason, there are many fewer foulouts.  I am not sure why.  Let’s look at the batter pool and then the adjusted numbers:

foul 1.00
s 1.01
ex 1.00
d/ex 1.00
t/ex 0.99
hr 0.98
bb 1.00
hp 0.99
so 0.99
roe 1.00

A pretty typical batter, a little short on HR and long on singles, a pool probably over-represented by the #2 batter in the order.  Here are the batter adjusted rates for this base/out state:

foul 0.94
s 1.04
ex 0.97
d/ex 1.00
t/ex 1.04
hr 0.83
bb 1.14
hp 1.4
so 0.87
roe1.21

OK, one more. Let’s compare that to a runner on second and 2 outs, where the batter is definitely not trying to move the runner over, the pitcher is not trying to strike out the batter (just get an out), although here he definitely does mind a walk, and the batter should be trying to get a hit to drive in the runner (as opposed to hitting an extra base hit or home run like he might with a runner on 1st or no one on) and definitely does not want to walk if he can help it.

foul 0.97
s 0.97
ex 1.04
d/ex 0.97
t/ex 1.25
hr 0.97
bb 1.5
hp 1
so 1.09
roe 0.92

Kind of an odd profile with 2 outs and a runner on 2nd.  First of all, I mispoke when I said that with runners on base there was always a high ROE rate.  With 2 outs, that is not usually the case, other than 2nd and 3rd for some reason.  Anyway, even though batters should be trying to get more singles and not try for the extra base hit, they get fewer singles and more extra base hits.  Fewer HR’s though, which is probably expected. In fact, with 2 outs, you see HR depressed across the board.  I guess that pitchers do not like to give up the HR with 2 outs and rightfully so.  So pitchers trump there.  Lots more BB as you would expect, but only average HP, which is odd, as in other base/out states, as I said, pitchers appear to control the HP and they tend to mimic the BB rate. Also, a lot of K for some reason.  Neither batters nor pitchers should be trying for the K.  Maybe because pitchers are throwing to the corners no minding a walk, batters are K’ing more. With one out (and a runner on second), we see high BB but not high K, so maybe batters are more impatient with 2 outs and swing more at the bad pitches.  We also see more triples relative to doubles.  I am not sure why. With 2 outs, you would expect to see batters not trying for the third when it is going to be close.  The only thing I can think of, which is probably true, is that with a runner on 2nd and 2 outs, the OF is playing a little close to cut down the runner on a hit, such that a ball in the gap or over their heads is more likely to be a triple. THat may be the reason we see more extra base hits in general and fewer singles, the exact opposite of what we would expect from the correct batter approach.

Let’s look at the pool of batters:

foul 1.00
s 1.00
ex 1.00
d/ex 1.00
t/ex 0.99
hr 1.03
bb 1.01
hp 1.01
so 1.01
roe 0.99

Fairly normal - a little heavy on the HR and walks.

So here are the overall rates for that base/out state, adjusted for the batters:

foul 0.97
s 0.97
ex 1.04
d/ex 0.97
t/ex 1.26
hr 0.94
bb 1.48
hp 0.99
so 1.08
roe 0.92

Next time I will look at the pitcher pools for each base/out state to see if that makes much of a difference.  That might make a big difference, but then again, you have more of a circular effect (although the effect could be in the opposite direction - for example, if a pitcher walks the bases loaded a lot - with the bases loaded, he will walk fewer than his overall walk rate). For example, with the bases loaded, you probably have a pool of pitchers who give up a lot of hits and walks.  But then again, their overall stats are influenced by the fact that they face a lot of batters with the bases loaded.

BTW, I am not the only person to come up with this, but in small samples, maybe even for a year or two. we probably should be adjusting batters lines for the base/out state they come up in.  For example, with the sluggers, their walk (non-IBB) totals will clearly be heavily influenced by how often they come to bat with a base open.