Avoid K's a Worthless Rating?

brain hemorrhage · 02-06-2005, 05:30 PM

Hello, I have been playing around with OOTP6 and it seems to me like Avoid Ks is a completely worthless rating that inflates the Blue and Gold Star Ratings of players.

Avoid Ks only seems to ever be useful if you have a really fast contact hitter. But then, if you subscribe to the Moneyball school of thinking, these players are severely overvalued in trades because they are a rare commodity. Not only does the sim appear to artificially inflate players Star Ratings across the board due to their ability to avoid Ks, but if a player is fast his Star Rating will be inflated even further.

On the other hand, I have seen remarkable value in Eye/Discipline. Players with a 100 (1-100 scale) Eye/Discipline seem to have their career productivity extend much longer. Eye/Discipline seems to be the most steadiest rating in old age and is much harder to drop than other ratings. Not only that but even as the player gets older he continues to get on base by walking. I had a 38yo 1B fall apart during his last season and by the end of the year he was batting below the Mendoza Line. Yet, he had a staggering 145 walks!!! Making him a very serviceable player due to a ridiculous OBP.

I've read a couple of Ratings studies on OOTP6 and there is not much talk about the correlation between Eye/Discipline vs. Avoid Ks and runs scored/ERA. What would be even more interesting is to see what the effect of speed and baserunning had as a variable.

NCBeachBum · 02-06-2005, 05:40 PM

I see this much differently than the majority of amateur sabermetricians out there who think K's don't matter for a hitter. If you believe in DIPS, and you're going to have a certain average batting average on balls put in play, then the higher the percentage of your AB's in which you place a ball in play, the higher your BA and OBP will be. Just as a pitcher who doesn't K many places his fate in the hands of his defense, a hitter who doesn't K a lot gives the other team's defense more chances to not get to the ball.

brain hemorrhage · 02-06-2005, 05:54 PM

That's fair but you are talking from a baseball theory view. I am speaking from a OOTP view. The statistical outputs for a guy with a 1 in avoid Ks is not that different from a guy with a 6 in avoid Ks in some cases... Obviously if you are a poor hitter and cannot avoid Ks, you will strike out a lot. However my issue is this;

Given 30 points on a 1-10 scale... how would you prefer to see the distribution of points work?

6 Contact
5 Gap
8 Power
10 Eye/Discipline
1 Avoid Ks ... this seems to be like an optimal setting. If you trade Eye/Discipline for Avoid Ks and instead have 1 Eye/Discipline and 10 Avoid Ks you will certainly test the defense a lot more, but nothing tests a defense like a baserunner coming home from third. First things first, get players on base. Then worry about finding holes in the defense.

Spielman · 02-06-2005, 06:01 PM

Quote:

Originally Posted by NCBeachBum

I see this much differently than the majority of amateur sabermetricians out there who think K's don't matter for a hitter. If you believe in DIPS, and you're going to have a certain average batting average on balls put in play, then the higher the percentage of your AB's in which you place a ball in play, the higher your BA and OBP will be. Just as a pitcher who doesn't K many places his fate in the hands of his defense, a hitter who doesn't K a lot gives the other team's defense more chances to not get to the ball.

#1 - Batters DO have some control over their BABIP, so a hitter who strikes out a lot can still have a decent batting average.

#2 - This means DIPS doesn't apply to hitters in the same way it does to pitchers.

#3 - It means the impact of K's isn't as high as you first might think.

#4 - Despite all this no "amateur sabermetrician" who knows anything is actually saying that strikeouts are meaningless in real live baseball. Clearly, they're a negative. It's just that they can be more than balanced out by many other factors which make a guy like Sammy Sosa a pretty ok player despite them.

-Spielman

lynchjm24 · 02-06-2005, 06:10 PM

I think the major point on strikeouts is that they are no different then a regular out. What you would pick up in 'productive outs', you lose in double plays.

If a guy hits .280/.350/.500 and strikeouts 120 times and a guy hits .280/.350/.500 and strikeouts 70 times, there really is no difference between these player's offensive contributions.

Crapshoot · 02-07-2005, 01:04 AM

K's matter amongst younger hitters as a sign of development- but beyond that, they're pretty much as harmufl as any other out. Remember, not putting the ball in plays limits your ability to cause more than 1 out- the AJ Pryszenski rule.

JDOldSchool · 02-07-2005, 05:28 AM

Quote:

Originally Posted by NCBeachBum

I see this much differently than the majority of amateur sabermetricians out there who think K's don't matter for a hitter. If you believe in DIPS, and you're going to have a certain average batting average on balls put in play, then the higher the percentage of your AB's in which you place a ball in play, the higher your BA and OBP will be. Just as a pitcher who doesn't K many places his fate in the hands of his defense, a hitter who doesn't K a lot gives the other team's defense more chances to not get to the ball.

K's are, on average, only a little bit worse than any other out. The slight difference is not enough to make a guy like Adam Dunn an unproductive player. A guy like Mike Cameron or Corey Patterson? That's probably a more debateable topic.

Fibonacci · 02-07-2005, 05:41 AM

Tee only time I really consider the K avoidance is in my number 2 batting slot. Even then, it's still far less important to me than eye/discipline.

swat-spas · 02-07-2005, 03:05 PM

In my test leagues for all-time players what I am seeing is that the players with higher avoid K ratings tend to 'underperform' their adjusted performance expectations. This is dependent on things like gap and power especially (good ratings here manage to eclipse the problem). They still get hits at a good clip, but without speed they don't amount to as much as their contact ratings suggest they should. I suppose the opposite end is someone who never strikes out but supposedly can't hit.. this person might see an improvement in batting average due to BABIP. In general though, stick to eye ratings. All this rating does is decrease Ks..in so doing it basically mates their contact rating to the BABIP setting for the league, which can have poor effect.

Ol'Drippy · 02-07-2005, 03:07 PM

Quote:

Originally Posted by Aadik

K's matter amongst younger hitters as a sign of development- but beyond that, they're pretty much as harmufl as any other out. Remember, not putting the ball in plays limits your ability to cause more than 1 out- the AJ Pryszenski rule.

I'm glad something double play related got named after him.

mrbill · 02-07-2005, 03:10 PM

Thats an interesting hypothesis, so you are claiming that someone with above-average Contact suffers with above-average avoiding Ks compared to someone with low avoiding Ks? And, this is true unless they have Ichiro-like speed, where the extra avoiding Ks ability helps?

I'm trying to see the connection between putting the ball in play more and having a lower average, its not like those Ks could be hits somehow, right?

Bluenoser · 02-07-2005, 03:13 PM

I kind of like the rating myself. I try to put a guy with a good avoid K rating batting 2nd in my line-up. He may not get a hit but he does tend to make contact with the ball, put it into play. Since I like the hit & run option, this is a good thing if my lead off hitter gets on.

canadiancreed · 02-07-2005, 03:21 PM

my understand of the avoid K rating is this, with all other ratings being equal:

a player of 1 avoid K rating will look like Pete Ingciviliga(sp?), hitting for power and contact, but striking out a ton.

a player of 10 avoid k will severly test the defense with line drives, grounders, fly balls, dependant on the pitchers GB rating. this forces the defense to earn the out, and hence more chance for getting on due to errors and scoring unearned runs.

crazyhorsejohnny · 02-07-2005, 03:24 PM

Quote:

Originally Posted by Spielman

It's just that they can be more than balanced out by many other factors which make a guy like Sammy Sosa a pretty ok player despite them.

Pretty OK player? I'd say a player who has hit 574 hrs in 16 seasons is better than OK, very good would be the term.

UngratefulDead · 02-07-2005, 05:07 PM

Quote:

Originally Posted by brain hemorrhage

6 Contact
5 Gap
8 Power
10 Eye/Discipline
1 Avoid Ks ...

http://www.baseball-reference.com/d/deerro01.shtml

Well.. less contact obviously. But close.

Raiderkenny · 02-07-2005, 11:11 PM

I believe that, with the new system, the probability for a hit is calculated after a strikeout is determined. Therefore, contact rating would determine the hitting percantage of balls put in play and not their final batting average. If that doesn't make enough sense then hopefully this following example will.

For example, you might have two players with a contact rating of 70/100. And let's say that this translates to an about an .333 percentage on balls put in play. Furthermore, player A would have an avoid K rating of 20/100 while player B would have an avoid K of 80/100. Finally, A subsequently strikes out 150 times over a season while B strikes out 50 times.

With both players having 600 ABs, player A would be putting the ball in play 450 times (since 150 of those ABs were strikeouts). With a hitting rate of .333, 150 of those would be hits. That would give him a final average of <b>.250</b>

Meanwhile, player B would be putting the ball in play 550 times. With the .333 rate, he would end up with about 183 hits. All in all, it would give him a much better looking batting average of <b>.305</b>

This is a crude example since it doesn't take many factors into account, but I think it displays that Avoid K's definitely are useful and can alter the batting average significantly. That being said, I still consider Avoid K's as the least most important rating behind the other 4.

Jayzone · 02-08-2005, 12:54 AM

Now, I don't claim that I know how the game works or know exactly which rating is for what. Well, I just want drop in and say something about contact rating. Contact rating actually does not correlate well with batter's BABIP. In contrast, contact rating acutally correlates a lot better with pure AVG. Well, I dug out some of the data that I have. From my one-year test league data, a correlation table between contact and both AVG and BABIP can be constructed.

Code:

Correlation Table between contact rating and the chosen rate stats

Sample Size    AVG  BABIP
        278  0.632  0.354
        189  0.645  0.348

Note:
1st row: data cutoff point at 100
2nd row: data cutoff point at 300

The first row comprise all players with (AB-SO-HR) > 100 and the second row comprise all players with (AB-SO-HR) > 300. So, the sample size if significant for rate stats.

As you can clearly see, contact rating acutally represent AVG significantly better than pure batter's BABIP so that I don't think BABIP is what contact represents. Oh, I actually did this for a few other test leagues and the results are more or less in agreement with this.

My guess on AvoidK is that its long-term impact is porbably very small but its short-term impact might be significant enough to warrant attention for short-term impact. What I meant is that two same ability-rated players with only vastly different avoidK rating will proably perform similarly at the end of a long season. However, if both players are picked to face a high/great strikeout pitcher during a game, then I think the better avoidK batter might perform noticeably better. Again, I don't know how OOTP produce/determine the stats outcome, the above is just my educated guess.

brain hemorrhage · 02-08-2005, 09:49 AM

If the GAP and other ratings in your study were not constants, your study isn't very meaningful...

Jayzone · 02-08-2005, 11:09 AM

Ok, you will have to remeber it was not a study to prove or disapprove anything. In fact, I have never said it's a study at all. I was merely trying to demonstrate what happenes in a typical OOTP league under default setting. If I were to do a study, I will, of course, do a controlled experiment and formulate everything but this is simply not that.

Just like I said before, I am just saying that contact is a better indicator for AVG than BABIP. Could contact rating represent BABIP? Possibly. I could neither prove or disapprove that by just relying on the data I shown. However, what I could show is that contact rating is a better indicator for AVG than BABIP under a real-league-environment. What I was set out to find out is about which rating is prmary influence for which rate stats. In other words, I was merely trying to see which rating is the best indicator for which rate stats (most apparent single rating to single rate stats relationship). There is an important distinction between what I did and trying to prove/disapprove how OOTP works internally. And I am not trying to do the latter.

02-07-2005, 01:04 AM	#6
Crapshoot Hall Of Famer Join Date: Feb 2002 Location: watching: DArwin's missing link in action Posts: 3,112	K's matter amongst younger hitters as a sign of development- but beyond that, they're pretty much as harmufl as any other out. Remember, not putting the ball in plays limits your ability to cause more than 1 out- the AJ Pryszenski rule. __________________ Senior Senor Member of the OOTP Boards Pittsburgh Playmates- OTBL

02-07-2005, 05:41 AM	#8
Fibonacci Minors (Triple A) Join Date: Sep 2004 Posts: 283	Tee only time I really consider the K avoidance is in my number 2 batting slot. Even then, it's still far less important to me than eye/discipline. __________________ It is not certain that everything is uncertain. - Blaise Pascal

02-07-2005, 03:21 PM	#13
canadiancreed Hall Of Famer Join Date: Aug 2004 Posts: 11,660	my understand of the avoid K rating is this, with all other ratings being equal: a player of 1 avoid K rating will look like Pete Ingciviliga(sp?), hitting for power and contact, but striking out a ton. a player of 10 avoid k will severly test the defense with line drives, grounders, fly balls, dependant on the pitchers GB rating. this forces the defense to earn the out, and hence more chance for getting on due to errors and scoring unearned runs. __________________ PT21 PT22

02-07-2005, 11:11 PM	#16
Raiderkenny Minors (Rookie Ball) Join Date: Jul 2002 Location: Ravena, NY Posts: 28	I believe that, with the new system, the probability for a hit is calculated after a strikeout is determined. Therefore, contact rating would determine the hitting percantage of balls put in play and not their final batting average. If that doesn't make enough sense then hopefully this following example will. For example, you might have two players with a contact rating of 70/100. And let's say that this translates to an about an .333 percentage on balls put in play. Furthermore, player A would have an avoid K rating of 20/100 while player B would have an avoid K of 80/100. Finally, A subsequently strikes out 150 times over a season while B strikes out 50 times. With both players having 600 ABs, player A would be putting the ball in play 450 times (since 150 of those ABs were strikeouts). With a hitting rate of .333, 150 of those would be hits. That would give him a final average of <b>.250</b> Meanwhile, player B would be putting the ball in play 550 times. With the .333 rate, he would end up with about 183 hits. All in all, it would give him a much better looking batting average of <b>.305</b> This is a crude example since it doesn't take many factors into account, but I think it displays that Avoid K's definitely are useful and can alter the batting average significantly. That being said, I still consider Avoid K's as the least most important rating behind the other 4. Last edited by Raiderkenny; 02-07-2005 at 11:14 PM.

02-08-2005, 12:54 AM	#17
Jayzone All Star Reserve Join Date: Apr 2004 Posts: 706	Contact rating actually correlates much better with AVG than batter's BABIP Now, I don't claim that I know how the game works or know exactly which rating is for what. Well, I just want drop in and say something about contact rating. Contact rating actually does not correlate well with batter's BABIP. In contrast, contact rating acutally correlates a lot better with pure AVG. Well, I dug out some of the data that I have. From my one-year test league data, a correlation table between contact and both AVG and BABIP can be constructed. Code: Correlation Table between contact rating and the chosen rate stats Sample Size AVG BABIP 278 0.632 0.354 189 0.645 0.348 Note: 1st row: data cutoff point at 100 2nd row: data cutoff point at 300 The first row comprise all players with (AB-SO-HR) > 100 and the second row comprise all players with (AB-SO-HR) > 300. So, the sample size if significant for rate stats. As you can clearly see, contact rating acutally represent AVG significantly better than pure batter's BABIP so that I don't think BABIP is what contact represents. Oh, I actually did this for a few other test leagues and the results are more or less in agreement with this. My guess on AvoidK is that its long-term impact is porbably very small but its short-term impact might be significant enough to warrant attention for short-term impact. What I meant is that two same ability-rated players with only vastly different avoidK rating will proably perform similarly at the end of a long season. However, if both players are picked to face a high/great strikeout pitcher during a game, then I think the better avoidK batter might perform noticeably better. Again, I don't know how OOTP produce/determine the stats outcome, the above is just my educated guess. Last edited by Jayzone; 02-08-2005 at 12:55 AM.

02-06-2005, 05:30 PM	#1
brain hemorrhage Minors (Rookie Ball) Join Date: Feb 2005 Posts: 37	Avoid K's a Worthless Rating? Hello, I have been playing around with OOTP6 and it seems to me like Avoid Ks is a completely worthless rating that inflates the Blue and Gold Star Ratings of players. Avoid Ks only seems to ever be useful if you have a really fast contact hitter. But then, if you subscribe to the Moneyball school of thinking, these players are severely overvalued in trades because they are a rare commodity. Not only does the sim appear to artificially inflate players Star Ratings across the board due to their ability to avoid Ks, but if a player is fast his Star Rating will be inflated even further. On the other hand, I have seen remarkable value in Eye/Discipline. Players with a 100 (1-100 scale) Eye/Discipline seem to have their career productivity extend much longer. Eye/Discipline seems to be the most steadiest rating in old age and is much harder to drop than other ratings. Not only that but even as the player gets older he continues to get on base by walking. I had a 38yo 1B fall apart during his last season and by the end of the year he was batting below the Mendoza Line. Yet, he had a staggering 145 walks!!! Making him a very serviceable player due to a ridiculous OBP. I've read a couple of Ratings studies on OOTP6 and there is not much talk about the correlation between Eye/Discipline vs. Avoid Ks and runs scored/ERA. What would be even more interesting is to see what the effect of speed and baserunning had as a variable.

02-06-2005, 05:40 PM	#2
NCBeachBum All Star Starter Join Date: Jul 2003 Location: Topsail Island, NC, USA Posts: 1,049	I see this much differently than the majority of amateur sabermetricians out there who think K's don't matter for a hitter. If you believe in DIPS, and you're going to have a certain average batting average on balls put in play, then the higher the percentage of your AB's in which you place a ball in play, the higher your BA and OBP will be. Just as a pitcher who doesn't K many places his fate in the hands of his defense, a hitter who doesn't K a lot gives the other team's defense more chances to not get to the ball.

02-06-2005, 05:54 PM	#3
brain hemorrhage Minors (Rookie Ball) Join Date: Feb 2005 Posts: 37	That's fair but you are talking from a baseball theory view. I am speaking from a OOTP view. The statistical outputs for a guy with a 1 in avoid Ks is not that different from a guy with a 6 in avoid Ks in some cases... Obviously if you are a poor hitter and cannot avoid Ks, you will strike out a lot. However my issue is this; Given 30 points on a 1-10 scale... how would you prefer to see the distribution of points work? 6 Contact 5 Gap 8 Power 10 Eye/Discipline 1 Avoid Ks ... this seems to be like an optimal setting. If you trade Eye/Discipline for Avoid Ks and instead have 1 Eye/Discipline and 10 Avoid Ks you will certainly test the defense a lot more, but nothing tests a defense like a baserunner coming home from third. First things first, get players on base. Then worry about finding holes in the defense.

02-06-2005, 06:10 PM	#5
lynchjm24 All Star Reserve Join Date: Jun 2002 Location: Hartford Posts: 978 Infractions: 0/1 (1)	I think the major point on strikeouts is that they are no different then a regular out. What you would pick up in 'productive outs', you lose in double plays. If a guy hits .280/.350/.500 and strikeouts 120 times and a guy hits .280/.350/.500 and strikeouts 70 times, there really is no difference between these player's offensive contributions.

02-07-2005, 03:05 PM	#9
swat-spas Minors (Single A) Join Date: Aug 2002 Posts: 53	In my test leagues for all-time players what I am seeing is that the players with higher avoid K ratings tend to 'underperform' their adjusted performance expectations. This is dependent on things like gap and power especially (good ratings here manage to eclipse the problem). They still get hits at a good clip, but without speed they don't amount to as much as their contact ratings suggest they should. I suppose the opposite end is someone who never strikes out but supposedly can't hit.. this person might see an improvement in batting average due to BABIP. In general though, stick to eye ratings. All this rating does is decrease Ks..in so doing it basically mates their contact rating to the BABIP setting for the league, which can have poor effect.

02-07-2005, 03:10 PM	#11
mrbill All Star Reserve Join Date: Jul 2003 Location: LA, CA Posts: 982	Thats an interesting hypothesis, so you are claiming that someone with above-average Contact suffers with above-average avoiding Ks compared to someone with low avoiding Ks? And, this is true unless they have Ichiro-like speed, where the extra avoiding Ks ability helps? I'm trying to see the connection between putting the ball in play more and having a lower average, its not like those Ks could be hits somehow, right?

02-07-2005, 03:13 PM	#12
Bluenoser Hall Of Famer Join Date: Mar 2002 Location: In The Moment Posts: 14,521	I kind of like the rating myself. I try to put a guy with a good avoid K rating batting 2nd in my line-up. He may not get a hit but he does tend to make contact with the ball, put it into play. Since I like the hit & run option, this is a good thing if my lead off hitter gets on.

02-08-2005, 09:49 AM	#18
brain hemorrhage Minors (Rookie Ball) Join Date: Feb 2005 Posts: 37	If the GAP and other ratings in your study were not constants, your study isn't very meaningful...

02-08-2005, 11:09 AM	#19
Jayzone All Star Reserve Join Date: Apr 2004 Posts: 706	Ok, you will have to remeber it was not a study to prove or disapprove anything. In fact, I have never said it's a study at all. I was merely trying to demonstrate what happenes in a typical OOTP league under default setting. If I were to do a study, I will, of course, do a controlled experiment and formulate everything but this is simply not that. Just like I said before, I am just saying that contact is a better indicator for AVG than BABIP. Could contact rating represent BABIP? Possibly. I could neither prove or disapprove that by just relying on the data I shown. However, what I could show is that contact rating is a better indicator for AVG than BABIP under a real-league-environment. What I was set out to find out is about which rating is prmary influence for which rate stats. In other words, I was merely trying to see which rating is the best indicator for which rate stats (most apparent single rating to single rate stats relationship). There is an important distinction between what I did and trying to prove/disapprove how OOTP works internally. And I am not trying to do the latter.