Performance with 2 strikes

[Scorekeeper]

On one of the baseball boards, the topic of 2 strike hitting came up. I’ve been in many a discussion about that over the years, and one thing that’s always bothered me, was how anyone managed to identify good 2 strike hitters to begin with. So, this time I decided to bite the bullet and see if I could come up with some way to quantify how hitters performed with 2 strikes on them.

Well, after experimenting a bit, I’ve managed to quantify how both our pitchers and our hitters REALLY deal with 2 strike counts to at least some degree.

http://www.infosports.com/scorekeeper/images/2strikesboth.pdf

The thing I see both intuitively obvious as well as proved true by the numbers is, hitters in general don’t fare very well with 2 strikes on them, and pitchers in general do very well when they get 2 strikes on hitters. Its intuitively obvious when one considers its only when there are 2 strikes that the additional possibility of making an out exists.

Rather than just throw out clichés with no proof other than “gut feelings” or “perceptions” about how hitters and pitchers compare when they are in 2 strike counts, at least now I have the ability to look at our pitchers and hitters real numbers. While I think its safe to say all hitters and pitchers can do better in those situations, with numbers in hand, those who need the most help can be easily identified and given more problem specific help. That should lead to much more efficient use of valuable coaching talent and time.

Does anyone out there have anything similar to use for FPSB? I’m pretty sure the same conversations go on about how player “A” is a great 2 strike hitter, or pitcher “Z” is murder on hitters when she gets 2 strikes on them. In BB there’s very few people who even have the capability to track such things, but I’m thinking that the higher evolved scorers in SB might have something on it.

 

[James Smith]

I think in this case the stats might lie. As a pitcher DD is up 0-2 because she is a better pitcher then the batter is, as a batter she is down 0-2 because the pitcher is better. Both favor the pitcher.

(if DD is pitching with an 0-2 count she will hit the batter her next pitch)

 

[Scorekeeper]

I think in this case the stats might lie. As a pitcher DD is up 0-2 because she is a better pitcher then the batter is, as a batter she is down 0-2 because the pitcher is better. Both favor the pitcher.

(if DD is pitching with an 0-2 count she will hit the batter her next pitch)

If valid, stats never lie. They can be misinterpreted, but they can’t lie.

I really don’t believe it’s a matter of whether the pitcher is relatively better than the batter or vice versa. Its very possible a lousy pitcher may well get up 0-2 against a great hitter, and its very possible a lousy hitter might never go 0-2 against the greatest pitcher that ever lived. But I’m wondering why you’ve zeroed in on 0-2. There are 4 possible 2 strike counts. 0-2, 1-2, 2-2, and 3-2, and there are 5 possibilities for the outcome of any pitch. It can be a ball or any 1 of 4 types of strikes.

Why would any pitcher purposely hit a batter when the count is 0-2?

 

[CoogansBluff]

A player's performance in 2-strike situations is good info, but I'd think you'd want to analyze it further. Other questions -

How much worse does the team or league as a whole perform w/ 2 strikes? Let's say performance drops 50 percent (.300 average becomes .200). Does this player drop more or less than average?

What percentage of at-bats wind up in 2-strike situations (league or team-wide)? How does the player compare in that regard?

There are several strategies that a player might employ to improve their hitting based on these stats, but those strategies would vary depending on all of those stats (not just one, like 2-strike BA). For example, strategy might be to be more or less selective w/ 2 strikes, but also might be to be more or less selective earlier in the count w/ the goal of avoiding 2-strike situations.

Seems like you'd need a lot of at-bats/data to make good conclusions.

btw, the best modern hitter w/ 2 strikes based on some quick research is probably Joe Mauer, a career .259 hitter w/ 2 strikes. Ichiro also hits .259 w/ 2 strikes, but he has a higher overall average, so Mauer wins my tiebreaker.

 

[SoCal_Dad]

Well, after experimenting a bit, I’ve managed to quantify how both our pitchers and our hitters REALLY deal with 2 strike counts to at least some degree.

http://www.infosports.com/scorekeeper/images/2strikesboth.pdf

Why do you refer to them as ABs rather than PAs?

Does the "Foul" column include both regular foul balls and foul tips? If so, that probably explains why the Ks column is higher than Call + Miss.

 

[Ken Krause]

Can't remember where I saw the stats, but I seem to recall that in MLB the worst count to have is 1-2. That's the one that results in the lowest batting average

 

[Scorekeeper]

A player's performance in 2-strike situations is good info, but I'd think you'd want to analyze it further. Other questions -

LOL! I always want analyze the numbers further.

How much worse does the team or league as a whole perform w/ 2 strikes? Let's say performance drops 50 percent (.300 average becomes .200). Does this player drop more or less than average?

Well, from the perspective of HS, any consideration of more than one team is virtually hopeless, at least at this point in time. We have 6 teams in our league, and as far as I know I’m the only one who can produce any kind of analysis based on 2 strike pitches. I wish there were more, but I’m afraid we’re still several years away from something like that.

However, there’s nothing wrong with seeing which players, if any, have performance which fails to meet the average. That’s something I was referring to when I noted that coaching efficiency could be improved.

What percentage of at-bats wind up in 2-strike situations (league or team-wide)? How does the player compare in that regard?

Again, information for the league is out of the question, but it wouldn’t be very difficult to get the percentage of at bats that go to 2 strikes vs the total number of at bats.

There are several strategies that a player might employ to improve their hitting based on these stats, but those strategies would vary depending on all of those stats (not just one, like 2-strike BA). For example, strategy might be to be more or less selective w/ 2 strikes, but also might be to be more or less selective earlier in the count w/ the goal of avoiding 2-strike situations.

Agreed. When I run numbers like this, I don’t do it to try to resolve anything really. What my main goal is, is to see if its even possible to get some numbers. These particular numbers identify players so they can be arranged from top to bottom as far as performance on 2 strikes measured by H+BB+HBP/AB w 2 strikes. Since I don’t chart pitches, I wouldn’t begin to know where to start to identify how selective anyone is. That’s something that to me is way beyond the current ability of data to show.

Seems like you'd need a lot of at-bats/data to make good conclusions.

No. As long as the data is valid, the conclusions are as well. What happens with larger samples is, the predictive nature of the numbers gets better. FI, a hitter who’s 1-2 is batting .500 no matter how you look at it. But whether or not he’s gonna continue to hit at that pace is a compete tossup, compared to the batter who’s 500-1000. They’re both batting .500, but the chances are far better that the 2nd will hit at that pace.

btw, the best modern hitter w/ 2 strikes based on some quick research is probably Joe Mauer, a career .259 hitter w/ 2 strikes. Ichiro also hits .259 w/ 2 strikes, but he has a higher overall average, so Mauer wins my tiebreaker.

As I was creating the code to run the report, I was acutely aware that numbers like the above don’t begin to tell the full story. I “almost” computed OBPs based not on the number of at bats, but rather on the number of pitches seen in a given count. FI, getting a hit on a 2-2 count that went B-B-S-S is different than one that went S-S-B-B OR B-B-S-S-S-S-S-S.

If the AB ends on the next pitch, in the 1st, only one 2 strike pitch was seen. In the 2nd 3 were seen, and in the 3rd, 5 were seen. That makes analysis very tricky.

 

[starsnuffer]

There's no such thing as a good 2 strike hitters. There's just hitters that miss chances until they have no rope left to hang from.

-W

 

[chinamigarden]

Gamechanger tracks at bats that go 3 or more pitches after the second strike. Then it calculates the percentages of 2 strike counts that go 3 or more additional pitches. Not sure how useful it is but at least it shows you who is prolonging the at bats. Gamechanger only does this for the hitters stats it does not track this for pitchers

 

[Scorekeeper]

Why do you refer to them as ABs rather than PAs?

Good question! I usually do differentiate, but to tell the truth, in OBR there really isn’t any difference. Actually, if I could remember to do it, I’d differentiate at bats by at bats that impact metrics required by the rules, and those that don’t. FI, for BA its really the number of hits divided by the number of at bats that impact BA. I.e., walks, HBPs, CI, sac flys, and and sac bunts have no impact.

I tried to do that for a while, but quite frankly it’s a real PITA. In this particular case it would prolly me more accurate to say PAs, but it didn’t even enter my mind.

Does the "Foul" column include both regular foul balls and foul tips? If so, that probably explains why the Ks column is higher than Call + Miss.

I know this might twist a few panties, but in my data a foul tip is not a foul ball, but rather a miss. The reason I do that is I don’t want to have to try to differentiate any more than I already do.

You really shouldn’t try to equate the columns on the right with the columns on the left. The columns indicating the count are the number of pitches seen with 2 strikes at that count. The number of types like call or foul are the number of times that happens with 2 strikes. The things like Ks or BBs are results of the entire at bat, and include all the pitches no matter what the count.

Yeah, I know it’s a bit confusing, but I was stumbling along in the development process, getting all kinds of different information with 2 strikes, and decided to put it all on 1 report rather than 3 like I typically would.

Nice observation though. It tells me you actually looked at the data, and that’s always appreciated. Since you did look, did you notice anything surprising between the “normal” stats and the 2 strike numbers?

What surprised me was, the player with the best OBP with 2 strikes had the worst “normal” OBP of all the regular starters, and the player who was the best of the regulars in normal OBP was one of the worst with 2 strikes.