Strikeouts & Winning... Does it matter?

[sveumrules]

No doubt the math is changing, as strikeouts continue to skyrocket contact becomes more scarce and therefore naturally more valuable. The question is, by how much?

 

I think there is definitely marginal value that extreme high contact hitters provide, assuming of course that they have something resembling a league average batting line, but at the end of the day I'm taking the guy with an 800 OPS & 22% K rate over the guy with the 750 OPS & 18% K rate assuming everything else is more or less equal.

 

But now lets say the 750 OPS guy has a K rate of 12% instead, maybe that's enough of a difference to make the 50 point OPS difference a wash. It's a similar link as the one between OBP & speed. Would you rather have a blazer with a 335 OBP or a base clogger with a 350 OBP?

[logan82]

All stats are is a way to judge a player's talent. Strikeouts don't correlate very well with whatever underlying skill that makes a hitter good or bad so they are not a good way to judge a hitter. They do however seem to correlate well with how good a pitcher is. My guess is pitchers with more strikeouts have better control or velocity making it tougher for hitters to hit the ball well off of them. The underlying skill for a hitter would probably be how well they can drive a ball which could actually show up as more strikeouts if they are more selective with what they swing at.

 

José Hernandez was an interesting case. He was with the Brewers for three years. The year he struck out the most was his most productive at the plate (188 K, .356 OBP, .478 SLG, .834 OPS). The year he struck out the least was his least productive (125K, .315 OBP, .372 SLG, .687 OPS).

This seems to be a pretty good example of what happens to a hitter when they try to strikeout less.

 

But now lets say the 750 OPS guy has a K rate of 12% instead, maybe that's enough of a difference to make the 50 point OPS difference a wash. It's a similar link as the one between OBP & speed. Would you rather have a blazer with a 335 OBP or a base clogger with a 350 OBP?

I think it would take more than 50 points of OPS. The difference in a strikeout vs a ball in play out is pretty small especially after you figure in extra double plays which are way worse than a strikeout.

[CheezWizHed]

Here is an article that breaks down why Ks mean different things for pitchers than they do for hitters...

 

http://www.beyondtheboxscore.com/2013/4/1/4165664/how-can-strikeouts-be-great-for-pitchers-but-not-that-bad-for-hitters

 

Money quote...

 

"The correlation between K% and wOBA for that same sample of batters is small at r = .12, suggesting that if anything wOBA increases by the slightest margin as K% increases. Not so bad for those hitters.

 

Pitcher's K% has a much stronger correlation to ERA at -.52, however, using a sample of 1071 pitchers from 2002-2012 with at least 150 IP for the season. Clearly, very good for pitchers."

 

Two problems that I have with this article:

1) The author spends no time defending the "Ks are good for pitching" argument and spends the lion share of data on the batter. I actually don't doubt that Ks don't correlate to poor offense. Also looking at all numbers quoted in this thread and its focused on the batter and nothing on the pitcher.

-> Please understand: My premise is that Ks have the same effect on the batter to score runs as it does the pitcher to prevent runs.

 

2) The one and only number he uses to defend the pitcher (Pitcher's K% has a much stronger correlation to ERA at -.52) is talking about EARNED runs. That is a problem. A win is not determined by an EARNED run, but ALL runs. Certainly it makes sense that the more Ks a pitcher has, the less he relies on his defense. But that isn't the point. The question is: Strikeouts and Winning. Winning is determined by ALL runs.

 

I think that math is "logical" by definition. It's common sense that isn't adding up. The discrepancy is definitely befuddling, but it is what it is. sveumrules has offered up some numbers.

 

Math is of course. Application of statistics can be very misleading (note I pointed out the discrepancy between ERA and RA above). Please be aware that I'm a software engineer and took enough math classes in college to get a minor in it. I'm not a "genius", but I'm no rube either. I've often wondered if this stuff was around when I was younger if I'd be working for the Brewers now.

[1992casey]

I think I'd rather correlate to runs too, but my gut says that it's likely OK to correlate to ERA as long as the sample is large enough.

[CheezWizHed]

I was back and forth on ERA vs RA also, but that is the main problem with the numbers being presented.

 

On one side, the batter's numbers are being filtered. No player with under 300 PA's is included in the data. So, you have automatically eliminated the worst players from the statistics. They do that with the premise of trying to look at "good" players, which is fine as long as you know the limitations of those statistics (remember, I said math was logical, application of statistics is not always logical).

 

But then you do the same thing for pitchers. We only look at ERA and not RA. Certainly that makes sense if you want to compare a singular pitcher and separate the effects of his defense. But if you want to look at the effect of Ks on offense, those numbers don't give us a complete pitcher.

 

So taking numbers that are "slightly off" on both sides and comparing them gives a conclusion that is slightly off.

 

Maybe to drive home my point... The author lists a table with "Run Values for Outs" in the batter section. A strikeout is worth -0.3 runs. That isn't calculated as -0.3 runs "for the batter". Its simply -0.3 runs scored. Its effective for both the pitcher and batter, right?

[Thurston Fluff]

If compare two players with the same power and on base skills, but one has far worse strikeout numbers than the other are they equal? Isn't it likely the player who strikes out the least will be marginally more productive than the player who has strikes out the most?

[logan82]

If compare two players with the same power and on base skills, but one has far worse strikeout numbers than the other are they equal? Isn't it likely the player who strikes out the least will be marginally more productive than the player who has strikes out the most?

Sure all else being equal I would take the guy who strikes out less, but strikeouts are far down the list of what I would look at. Probably the last thing.

 

Off the top of my head the order of things I would look at.(RBIs do not make the list)

wOBA

wrc+

OPS

OBP, SLG, BA how they compare or where a player gets their offensive value. The combination will give a good idea of walk rates, whether they rely on balls in play, how much power they have.

defense

Baserunning (mostly SB vs CS)

HR

Strikeouts

[Ennder]

Players who strike out too much only stick around the league if they have other tools that make them valuable. Pitchers who don't strike out a lot of hitters only stick around in the league if they have other tools that really make them valuable (low BB%, heavy GB% etc).

 

This tends to weed out the poor high K hitters naturally and it weeds out the poor low K pitchers as well. So if you take the star players out of the equation for a moment you are not going to survive as a mediocre hitter with a ton of Ks because it will drive your OBP down too much and you aren't going to survive as a mediocre pitcher without a ton of Ks. This skews things towards Ks not mattering for hitters as much as it does for pitchers.

[Thurston Fluff]

If compare two players with the same power and on base skills, but one has far worse strikeout numbers than the other are they equal? Isn't it likely the player who strikes out the least will be marginally more productive than the player who has strikes out the most?

Sure all else being equal I would take the guy who strikes out less, but strikeouts are far down the list of what I would look at. Probably the last thing.

 

Off the top of my head the order of things I would look at.(RBIs do not make the list)

wOBA

wrc+

OPS

OBP, SLG, BA how they compare or where a player gets their offensive value. The combination will give a good idea of walk rates, whether they rely on balls in play, how much power they have.

defense

Baserunning (mostly SB vs CS)

HR

Strikeouts

 

So it does matter, just not much. That is not the same as it doesn't matter at all. It is a weakness. Which leads me to another point. If a team completely ignores strikeouts they can end up with a lot of players who strikeout a lot. A few free swingers may not hurt but does that mean a team full of them doesn't? Is it possible that small weakness in one or two players gets amplified when all the players share that weakness? Has that been accounted for in the studies? I do not know but I am not sure anyone else does either. I ask because it seems relevant when rebuilding. Not to mention I think some of that was going on with the team the past couple seasons.

[logan82]

Sure they matter but not enough that they should come into a conversation more than once in a great while. I doubt you are going to find two players very often that it will come down to strikeouts.

 

I also forgot to add position ahead of strikeouts.

 

I believe a few years ago Melvin said they would have to turn 100 strikeouts into in play outs to gain one run. There are so many things that matter more than strikeouts.

 

A few free swingers may not hurt but does that mean a team full of them doesn't?

Hard to say but I would guess no as long as they do other things well. It would be a frustrating team to watch.

[stalton]

Imagine 8 Russell 3TO Branyan's in the lineup.

[DuWayne Steurer]

Imagine 8 Russell 3TO Branyan's in the lineup.

 

 

If you have 8 batters with a .330 OBP, and hitting a lot of homeruns, you'll score a ton of runs.

[JimH5]

Imagine 8 Russell 3TO Branyan's in the lineup.

 

 

If you have 8 batters with a .330 OBP, and hitting a lot of homeruns, you'll score a ton of runs.

 

 

True.

 

Branyan's career stats are:

 

3398 PA

2934 AB

682 H

403 BB

1118 K

192 HR

.232 BA

 

Converted to a pitching line, that works out to:

 

750.2 IP

8.18 H/9

4.83 BB/9

13.40 K/9

2.30 HR/9

 

A pitcher who allowed those rate stats would never be around long enough to log 750.2 IP.

[CheezWizHed]

Of course, we would need 15 Russells just to keep 8 of them healthy enough to be in the lineup at one time.

[topper09er]

The missing link that is hard to measure is the trade-off between BABIP and the added OPS you get from taking more pitches and swinging harder. A K is the same as another out most of the time, but a ball in play is better than a K. You can't assume that every strikeout would just have been a ground out or something, so it is not equal to "another out". The league average BABIP was .302 last year, so if there were no strikeouts we could assume about 30% of those ABs would have resulted in a hit. So since we know K's do not directly correlate with crappy players, there must be a benefit to striking out. The benefit is obvious, if you K you take more pitches which means your chances to walk increases, and if you K you take bigger swings which increases your slugging if you make contact. So its really a balance between OBP, SLG and K. If you are good at all of those you are a HOFer because you dont have the missed BABIP penalty of striking out but can still maintain the high OPS. For regular players you just need to look at the OPS and ignore the Ks, it doesnt matter how they get there.

[DuWayne Steurer]

The missing link that is hard to measure is the trade-off between BABIP and the added OPS you get from taking more pitches and swinging harder. A K is the same as another out most of the time, but a ball in play is better than a K. You can't assume that every strikeout would just have been a ground out or something, so it is not equal to "another out". The league average BABIP was .302 last year, so if there were no strikeouts we could assume about 30% of those ABs would have resulted in a hit. So since we know K's do not directly correlate with crappy players, there must be a benefit to striking out. The benefit is obvious, if you K you take more pitches which means your chances to walk increases, and if you K you take bigger swings which increases your slugging if you make contact. So its really a balance between OBP, SLG and K. If you are good at all of those you are a HOFer because you dont have the missed BABIP penalty of striking out but can still maintain the high OPS. For regular players you just need to look at the OPS and ignore the Ks, it doesnt matter how they get there.

 

 

I think, for some people, and I'm not saying anyone in particular, they just say "if you strike out less, that means those balls in play will be x% hits and homeruns and doubles" or so on. I think there's a fallacy to that thinking. Someone else mentioned that the season Jose K struck out 180+ times was his best offensive season. I'm not saying he had his best offensive season BECAUSE he struck out 180+ times, but I think there is causation between the offensive approach you have at the plate and the results.

 

If you take a guy like Jose Hernandez and say "we want you to choke up, protect the plate, and make more contact", what do you end up with? Would you then get a .250 hitter with 25 doubles, 7 homeruns, and a .650 OPS, who doesn't take walks, and only strikes out 75 times? What benefit is there to that?

 

Apparently, that's exactly what Minnesota tried to do with Carlos Gomez. They wanted him to bunt, make contact, and use his speed to get on base with ground balls and such. I don't know how much truth there is to that. But if that's the case, it certainly wasn't working.

[topper09er]

Someone else mentioned that the season Jose K struck out 180+ times was his best offensive season. I'm not saying he had his best offensive season BECAUSE he struck out 180+ times, but I think there is causation between the offensive approach you have at the plate and the results.

 

If you take a guy like Jose Hernandez and say "we want you to choke up, protect the plate, and make more contact", what do you end up with? Would you then get a .250 hitter with 25 doubles, 7 homeruns, and a .650 OPS, who doesn't take walks, and only strikes out 75 times? What benefit is there to that?

 

Ideally what you want is for a player to choke up, protect the plate, and make more contact WITH 2 strikes. Not always. Nowadays every player takes the same approach every pitch (looking at you Rickie Weeks as exhibit A) What they should do is take their normal huge swings until they get to 2 strikes and then protect, unless they are are able to put up decent numbers with 2 strikes using their normal approach. These are usually all-star type players however.

 

In 2002, Jose Hernandez's best year when he struck out 188 times, his 2 strike line was .176/.219/.260/.479. 7 2B, 1 3B, 6 HR. That OPS is 24% below the league average, including pitchers hitting numbers. So it looks like JH should have choked up with 2 strikes and he would have had a better season, but only with 2 strikes so he would not have had to sacrifice the other power stats he got.

 

Take Mark Reynolds in 2009, 44 HR with 223 K. He was able to OPS just 6 points below the league average with 2 strikes while adding 14 2 strike HRs. So he would be an example where you probably would not need to adjust the approach with 2 strikes (except for key situations like R3 < 2 O but that is another topic) since he doesnt really suffer as much taking a huge swing with 2 strikes.

[Thurston Fluff]

Situational hitting as described by Topper09er is a skill that needs to be learned. If you go by the mantra strikeouts don't matter then it seems unlikely you would bother to learn that. It's also noteworthy that Hernandez's best year was below average league wide. Looking at how many k's one has in a given year and determining they don't matter because that one person's best season came when he had the most strikeouts misses the fact that he was always a strikeout prone player who was always below average.

[1992casey]

It's also noteworthy that Hernandez's best year was below average league wide

 

You've lost me here. Below average based on what?

 

José K, 2002: .356 OBP, .478 SLG, .834 OPS

NL average, 2002: .331 OBP, .410 SLG, .741 OPS

MLB average, 2002: .331 OBP, .417 SLG, .748 OPS

[Thurston Fluff]

It's also noteworthy that Hernandez's best year was below average league wide

 

You've lost me here. Below average based on what?

 

José K, 2002: .356 OBP, .478 SLG, .834 OPS

NL average, 2002: .331 OBP, .410 SLG, .741 OPS

MLB average, 2002: .331 OBP, .417 SLG, .748 OPS

 

My bad I didn't notice Topper was talking strictly about two strike counts. I think the point remains that not thinking strikeouts matter can lead to not learning how to cut down on them at the wrong time. Jose Hernandez' best season being the one with the most K's notwithstanding. He was always a strikeout hitter. The fact he had more one season than another most likely random chance. He accidentally hit a few more than he missed one season over another. I think it would have served him better over his career had he learned situational hitting better.

Maybe we are talking past each other a bit as well. I think cutting down on strikeouts makes a specific player better than they would have been. If a skill is not learned because of the concept that strikeouts don't matter then I think strikeouts matter. If only in the sense that the specific player is not as good as he could have been. If the conversation is about whether strikeout hitters as a whole hurts an offense then it's probably less of an issue.

[logan82]

Strikeouts as a tool for evaluation are not useful.

 

Whether a hitter could improve by being a situational hitter is definitely questionable. It could result in bad contact and more outs if a player adjusts their approach. maybe hitter do adjust and the results are bad because they did adjust.

[CheezWizHed]

I think, for some people, and I'm not saying anyone in particular, they just say "if you strike out less, that means those balls in play will be x% hits and homeruns and doubles" or so on. I think there's a fallacy to that thinking.

 

Back a couple pages, in that link talking about "Pitcher Ks=good. Batter Ks=doesn't matter", one good point he made was that some players strikeout because they swing hard. But they also get more 2Bs, 3B, and HRs because they swing hard also. So quite the opposite of that thinking.

 

There are some players that strike out a lot AND don't hit many XBHs. Those are the people to get rid of. But of you can strikeout but balance that with more XBHs, then more power to you (pun intended).

[danzig6767]

Players who strike out too much only stick around the league if they have other tools that make them valuable. Pitchers who don't strike out a lot of hitters only stick around in the league if they have other tools that really make them valuable (low BB%, heavy GB% etc).

 

This tends to weed out the poor high K hitters naturally and it weeds out the poor low K pitchers as well. So if you take the star players out of the equation for a moment you are not going to survive as a mediocre hitter with a ton of Ks because it will drive your OBP down too much and you aren't going to survive as a mediocre pitcher without a ton of Ks. This skews things towards Ks not mattering for hitters as much as it does for pitchers.

Yea, there are some hitters who strikeout a lot partly because they tend to take more pitches than average, so they don't swing often at boarderline strikes. This leads to both high walk and K totals, kinda like Dunn and Branyan type of hitters. People often think they K so much just because of how frequently they swing and miss, but a big factor is also that they generally avoid swinging at hard to hit strikes, so it results in lots of say 2-2 or 3-2 counts where a walk or K ends up happening.

 

That's much different than a total free swinger who K's a lot, but doesn't walk much either.