• Welcome to OGBoards 10.0, keep in mind that we will be making LOTS of changes to smooth out the experience here and make it as close as possible functionally to the old software, but feel free to drop suggestions or requests in the Tech Support subforum!

Mike Trout on pace for greatest season in MLB history

i’m actually guessing there is something to batter-pitcher matchups but the size of the effect is pretty small and thus impossible to measure given the sample sizes we are dealing with. so even if there are positive matchups, we can’t look at the data to find them — would have to be based on scouting / intuition. which is also where we most often are wildly wrong due to the subjectivity.
 
DistrictDeacon said:
And this is where you’re wrong. It’s a fact that most people DO regress to their averages. So it would be unwise and illogical to predict that a player would continue perform above average. Even though a small number beat the odds and do so.

That’s based on looking at every player and every at bat in MLB history. You can’t get a better sample size than that.
Click to expand...

READING IS FUNDAMENTAL!!!!

We AGREE on the fact that MOST are in that range. I've said the following (and more):

"I do understand what you say and agree a vast majority of the time. The stubborn one is you who can't admit that there are exceptions. "

"MOST is not ALL. I NEVER have discussed MOST. To everyone else here MOST EQUALS ALL.

I fully understand that the odds against a player owning or being owned is high. The others REFUSE that they exist AT ALL. That is irrational. "

But you and others say it's wrong to say the outliers are outliers and DO EXIST.

It's 100% WRONG to say they don't exist.

It's 100% wrong to say you can't use their ACTUAL results to say they will continue after they have reached critical mass.

You won't admit that we AGREE on well over 90% (probably over 95%) of players. I have NEVER disputed this.
 
RJKarl said:
READING IS FUNDAMENTAL!!!!

We AGREE on the fact that MOST are in that range. I've said the following (and more):

"I do understand what you say and agree a vast majority of the time. The stubborn one is you who can't admit that there are exceptions. "

"MOST is not ALL. I NEVER have discussed MOST. To everyone else here MOST EQUALS ALL.

I fully understand that the odds against a player owning or being owned is high. The others REFUSE that they exist AT ALL. That is irrational. "

But you and others say it's wrong to say the outliers are outliers and DO EXIST.

It's 100% WRONG to say they don't exist.

It's 100% wrong to say you can't use their ACTUAL results to say they will continue after they have reached critical mass.

You won't admit that we AGREE on well over 90% (probably over 95%) of players. I have NEVER disputed this.
Click to expand...

You do understand that when I said “a small number of players beat the odds” that means I know there are outliers right?

What the study says is that it is much more likely that those outliers will regress to the average than to remain outliers. Just because a few remained outliers doesn’t mean you can predict that most will.

And therefore it would be unwise to base your decision making on past pitcher batter matchups. You’re better off just playing your better overall hitter.
 
Exactly how many pitcher-batter matchups is RJ talking about here? 2 or 3 ?
 
DistrictDeacon said:
You do understand that when I said “a small number of players beat the odds” that means I know there are outliers right?

What the study says is that it is much more likely that those outliers will regress to the average than to remain outliers. Just because a few remained outliers doesn’t mean you can predict that most will.

And therefore it would be unwise to base your decision making on past pitcher batter matchups. You’re better off just playing your better overall hitter.
Click to expand...

I've only discussed the outliers. In less than two minutes of looking I found over half a dozen players whose entire careers prove that you can predict that a few will remain outliers and thus won't improve or regress to their overall averages.

To you, and especially MHB (and Chris), these don't exist. They do and should be handled as such.

But I'm RJ. Therefore you and others are free to say I'm wrong and rail against actual results rather than predictions.
 
I can't edit the above post to exclude "To You" before MHB.
 
RJKarl said:
I've only discussed the outliers. In less than two minutes of looking I found over half a dozen players whose entire careers prove that you can predict that a few will remain outliers and thus won't improve or regress to their overall averages.

To you, and especially MHB (and Chris), these don't exist. They do and should be handled as such.

But I'm RJ. Therefore you and others are free to say I'm wrong and rail against actual results rather than predictions.
Click to expand...

You can predict that A few will remain outliers. You just can’t predict WHO those few will be.

So therefore you can never know if an outlier will continue to be one. Or if they’ll go back to average.

So just play your player.
 
ChrisL68 said:
Because they are outliers. For every example you selectively post, there are multiple examples of guys who had good averages against pitchers after 35-40 plate appearances that regressed to the mean with more AB. And likely the same thing would happen with these guys the more AB that they had.
Click to expand...

I assume this means that players will progress to the mean as well. If a guy has a lousy average after 35-40 plate appearances but is a career .280 hitter, he wil progress to that .280 mean. Correct?
 
Shorty said:
I assume this means that players will progress to the mean as well. If a guy has a lousy average after 35-40 plate appearances but is a career .280 hitter, he wil progress to that .280 mean. Correct?
Click to expand...

yes
 
As I've said, if a player has 40 PAs against a pitcher or vice versa, you have enough data. Again, if a guy has a .325 career OBP and .350 vs. X pitcher, it's irrelevant. If he has a .450 or .500, there is a reason to use him. If a pitcher has held a player to a .150 OBP, I'll put that pitcher in to face that guy. If Manny is held powerless against a pitcher, it's a good to give him a rest.

As to a single PA, all things being relatively equal of course you go with the better player. But if I'm Danny Ozark and have the choice of whether I play the far superior Richie Hebner at 1B in a game versus Tom Seaver or Tommy Hutton. It's a no brainer. Richie Hebner gets the day off.

But MHB and Chris deny outliers exist. That was my point and that point was clearly repeated over and over again.
 
Shorty said:
I assume this means that players will progress to the mean as well. If a guy has a lousy average after 35-40 plate appearances but is a career .280 hitter, he wil progress to that .280 mean. Correct?
Click to expand...

In most cases, you'd be right, but there are outliers regardless of what Chris and MHB say.
 
That's better than most hit against Verlander. Of course, he plays. If he's hitting .160 (and isn't our 3/4/5 hitter), I give him the night off.
 
Shorty said:
Makes sense. So here is a scenario. Player A is a career .280 hitter, but after 40 ABs against Justin Verlander, he is hitting .240 against him. Does that mean given enough ABs, player A will eventually hit .280 against Verlander?
Click to expand...

Not necessarily if Verlander is better than the average pitcher player A faces over his career. What we are saying is that the .280 career average (or average over last 1500 ABs) is more statistically relevant to his next AB against Verlander than the .240 in his previous 40 ABs.
 
Shorty said:
Makes sense. So here is a scenario. Player A is a career .280 hitter, but after 40 ABs against Justin Verlander, he is hitting .240 against him. Does that mean given enough ABs, player A will eventually hit .280 against Verlander?
Click to expand...

No, he would be a 280 hitter against the average MLB pitcher. The mean against Verlander should be lower.
 
RJKarl said:
As I've said, if a player has 40 PAs against a pitcher or vice versa, you have enough data. Again, if a guy has a .325 career OBP and .350 vs. X pitcher, it's irrelevant. If he has a .450 or .500, there is a reason to use him. If a pitcher has held a player to a .150 OBP, I'll put that pitcher in to face that guy. If Manny is held powerless against a pitcher, it's a good to give him a rest.

I have posted multiple studies that disprove this.

As to a single PA, all things being relatively equal of course you go with the better player. But if I'm Danny Ozark and have the choice of whether I play the far superior Richie Hebner at 1B in a game versus Tom Seaver or Tommy Hutton. It's a no brainer. Richie Hebner gets the day off.

But MHB and Chris deny outliers exist. That was my point and that point was clearly repeated over and over again.

I don't deny outliers exist. I deny that they have any statistical value.
Click to expand...
 
MHBDemon said:
Not necessarily if Verlander is better than the average pitcher player A faces over his career. What we are saying is that the .280 career average (or average over last 1500 ABs) is more statistically relevant to his next AB against Verlander than the .240 in his previous 40 ABs.
Click to expand...

Not if the league is hitting 40 points lower against Verlander than the avaerage big league pitcher. Then player A’s .240 average is more statistically relevant in his next at bat against Verlander. I agree that once a player is an established player, he will progress or regress to his statistical mean against the LEAGUE. But not necessarily against a particular pitcher. That assumes all pitchers are the same.
 
Also, stop saying "read the linked studies!!1!" like they are scientific studies done by real scientists or even mathematicians. These are just made up blog boy exercises.

Their hope is that they make up a new made up stat that everyone likes and they get to move out of their aunt's basement.
 
You must log in or register to reply here.
Back
Top