Small sample sizes

[Frobby]

Mostly, this is just an excuse for me to post this very funny tweet I found from Tom Tango:

But, also just to say that I really don’t worry about individual players’ stats until the end of May.  Anyone can have a hot or cold month.   If it’s two months in a row, I start to notice.  And five days is just about meaningless, especially when judging veteran players with a track record of several years.   It’s not much more meaningful for young players (arguably, it’s less meaningful), but since there’s no major league track record to go on, it’s more tempting to extrapolate, even though the error bars are enormous.   

So in my view, the only stat that’s of much importance in the first two months are the team’s wins and losses, since you can’t take them back.  
 

[Pickles]

Has Frobby started to use Twitter?  That's two tweets posted today.  I fear for your mental health.

To the thread, I was actually just thinking about this last night.  I hate to pull up baseball reference home page right now because there's nothing to look at besides a bunch of numbers that have very little meaning.

I'll start to put some weight into the numbers after about a month.

[Sports Guy]

Yes but we have so much talent in the minors that if guys are sucking for 4-6 weeks, you may have to pull the plug on them quicker than normal.

[Frobby]

8 minutes ago, Pickles said:

Has Frobby started to use Twitter?  That's two tweets posted today.  I fear for your mental health.

Not really.  I’m not a member.  A while back, I’d sent Tom Tango (who’s a registered member here) a PM asking a question about why catch probability isn’t shown on the GameCast feed, and he responded to me with a link to a tweet on that subject.  I went to look at it today and ended up scrolling through various other things Tango had posted.  

By the way, Elon Musk has set up “X” to be virtually unusable for non-members, as posts don’t appear in any chronological order when a non-member visits a member’s page.  So I might see a tweet from Tango from 2017 right next to one from 2023.   If he thinks that’s going to encourage me to join to avoid that problem, he’s 100% wrong.  It just further alienates me. 
 

[Pickles]

11 minutes ago, Sports Guy said:

Yes but we have so much talent in the minors that if guys are sucking for 4-6 weeks, you may have to pull the plug on them quicker than normal.

I"ve seen little to suggest this org won't make quick adjustments.

Yes, are they going to give a guy like O'Hearn some rope after his performance and contributions last year?

Sure.  But if he's hitting .220 in May, he will be losing at-bats.

I think a great example is Mullins in 2022.  He was coming off his breakout 30-30 season; he was about the brightest light in a dreadful period.

And they didn't hesitate to both move him down in the order and to curtail his playing time against lefties when he struggled in 2022.

[Pickles]

1 minute ago, Frobby said:

Not really.  I’m not a member.  A while back, I’d sent Tom Tango (who’s a registered member here) a PM asking a question about why catch probability isn’t shown on the GameCast feed, and he responded to me with a link to a tweet on that subject.  I went to look at it today and ended up scrolling through various other things Tango had posted.  

By the way, Elon Musk has set up “X” to be virtually unusable for non-members, as posts don’t appear in any chronological order when a non-member visits a member’s page.  So I might see a tweet from Tango from 2017 right next to one from 2023.   If he thinks that’s going to encourage me to join to avoid that problem, he’s 100% wrong.  It just further alienates me. 
 

I didn't even know you could view twitter as a non-member.  Whenever I visit the homepage, it asks me to make an account.  Though I do seem to be able to view Twitter links that people post.

Just a friendly suggestion: 2024 might not be the year to join Twitter.

[Sports Guy]

12 minutes ago, Pickles said:

I"ve seen little to suggest this org won't make quick adjustments.

Yes, are they going to give a guy like O'Hearn some rope after his performance and contributions last year?

Sure.  But if he's hitting .220 in May, he will be losing at-bats.

I think a great example is Mullins in 2022.  He was coming off his breakout 30-30 season; he was about the brightest light in a dreadful period.

And they didn't hesitate to both move him down in the order and to curtail his playing time against lefties when he struggled in 2022.

But the thing is, to do this, players will have to go. Unless we have injuries, you aren’t sending guys to the minors that we have..unless we DFA them and they clear waivers.

[Frobby]

4 minutes ago, Pickles said:

I didn't even know you could view twitter as a non-member.  Whenever I visit the homepage, it asks me to make an account.  Though I do seem to be able to view Twitter links that people post.

Just a friendly suggestion: 2024 might not be the year to join Twitter.

No joke!  And no, I can’t get into Twitter via the home page as a non-member.  But I can look at a tweet embedded here and scroll on that person’s page and look at random chronological old posts from that person.  Not something I typically do, but I did today.  

[Pickles]

1 minute ago, Sports Guy said:

But the thing is, to do this, players will have to go. Unless we have injuries, you aren’t sending guys to the minors that we have..unless we DFA them and they clear waivers.

Kemp is not long for the roster one way or another.  There will be injuries.

Between injuries and non-performance I expect to need 2-3 bodies this season.

It's possible they DFA a guy like O'Hearn.  I wouldn't expect them to cut much deeper than that, just reallocate playing time.

[Say O!]

For those who want to go down the stats nerd rabbit hole — FanGraphs library piece on when do metrics stabilize (with link outs to other site pieces on reliability).

I think many would be shocked at how many ABs it takes for metrics to stabilize.

“Stabilization” Points for Offense Statistics:

• 60 PA: Strikeout rate
• 120 PA: Walk rate
• 240 PA: HBP rate
• 290 PA: Single rate
• 1610 PA: XBH rate
• 170 PA: HR rate
• 910 AB: AVG
• 460 PA: OBP
• 320 AB: SLG
• 160 AB: ISO
• 80 BIP: GB rate
• 80 BIP: FB rate
• 600 BIP: LD rate
• 50 FBs: HR per FB
• 820 BIP: BABIP

“Stabilization” Points for Pitching Statistics:

• 70 BF: Strikeout rate
• 170 BF: Walk rate
• 640 BF: HBP rate
• 670 BF: Single rate
• 1450 BF: XBH rate
• 1320 BF: HR rate
• 630 BF: AVG
• 540 BF: OBP
• 550 AB: SLG
• 630 AB: ISO
• 70 BIP: GB rate
• 70 BIP: FB rate
• 650 BIP: LD rate
• 400 FB: HR per FB
• 2000 BIP: BABIP

[Frobby]

14 minutes ago, Say O! said:

For those who want to go down the stats nerd rabbit hole — FanGraphs library piece on when do metrics stabilize (with link outs to other site pieces on reliability).

I think many would be shocked at how many ABs it takes for metrics to stabilize.

“Stabilization” Points for Offense Statistics:

• 60 PA: Strikeout rate
• 120 PA: Walk rate
• 240 PA: HBP rate
• 290 PA: Single rate
• 1610 PA: XBH rate
• 170 PA: HR rate
• 910 AB: AVG
• 460 PA: OBP
• 320 AB: SLG
• 160 AB: ISO
• 80 BIP: GB rate
• 80 BIP: FB rate
• 600 BIP: LD rate
• 50 FBs: HR per FB
• 820 BIP: BABIP

“Stabilization” Points for Pitching Statistics:

• 70 BF: Strikeout rate
• 170 BF: Walk rate
• 640 BF: HBP rate
• 670 BF: Single rate
• 1450 BF: XBH rate
• 1320 BF: HR rate
• 630 BF: AVG
• 540 BF: OBP
• 550 AB: SLG
• 630 AB: ISO
• 70 BIP: GB rate
• 70 BIP: FB rate
• 650 BIP: LD rate
• 400 FB: HR per FB
• 2000 BIP: BABIP

I’ve seen these kind of numbers, but I’m not really sure what “stabilize” really means in this context.   

[spleen1015]

19 minutes ago, Say O! said:

For those who want to go down the stats nerd rabbit hole — FanGraphs library piece on when do metrics stabilize (with link outs to other site pieces on reliability).

I think many would be shocked at how many ABs it takes for metrics to stabilize.

“Stabilization” Points for Offense Statistics:

• 60 PA: Strikeout rate
• 120 PA: Walk rate
• 240 PA: HBP rate
• 290 PA: Single rate
• 1610 PA: XBH rate
• 170 PA: HR rate
• 910 AB: AVG
• 460 PA: OBP
• 320 AB: SLG
• 160 AB: ISO
• 80 BIP: GB rate
• 80 BIP: FB rate
• 600 BIP: LD rate
• 50 FBs: HR per FB
• 820 BIP: BABIP

“Stabilization” Points for Pitching Statistics:

• 70 BF: Strikeout rate
• 170 BF: Walk rate
• 640 BF: HBP rate
• 670 BF: Single rate
• 1450 BF: XBH rate
• 1320 BF: HR rate
• 630 BF: AVG
• 540 BF: OBP
• 550 AB: SLG
• 630 AB: ISO
• 70 BIP: GB rate
• 70 BIP: FB rate
• 650 BIP: LD rate
• 400 FB: HR per FB
• 2000 BIP: BABIP

[emmett16]

44 minutes ago, Say O! said:

For those who want to go down the stats nerd rabbit hole — FanGraphs library piece on when do metrics stabilize (with link outs to other site pieces on reliability).

I think many would be shocked at how many ABs it takes for metrics to stabilize.

“Stabilization” Points for Offense Statistics:

• 60 PA: Strikeout rate
• 120 PA: Walk rate
• 240 PA: HBP rate
• 290 PA: Single rate
• 1610 PA: XBH rate
• 170 PA: HR rate
• 910 AB: AVG
• 460 PA: OBP
• 320 AB: SLG
• 160 AB: ISO
• 80 BIP: GB rate
• 80 BIP: FB rate
• 600 BIP: LD rate
• 50 FBs: HR per FB
• 820 BIP: BABIP

“Stabilization” Points for Pitching Statistics:

• 70 BF: Strikeout rate
• 170 BF: Walk rate
• 640 BF: HBP rate
• 670 BF: Single rate
• 1450 BF: XBH rate
• 1320 BF: HR rate
• 630 BF: AVG
• 540 BF: OBP
• 550 AB: SLG
• 630 AB: ISO
• 70 BIP: GB rate
• 70 BIP: FB rate
• 650 BIP: LD rate
• 400 FB: HR per FB
• 2000 BIP: BABIP

 

For one, barrels (for hitters) tend to stabilize more quickly than do outcome-based stats. We’ve seen before that exit velocity (one of the ingredients in the barrel recipe) tends to stabilize fairly quickly for hitters. “Barrel rate” (barrels/balls hit fair) is similarly quick to stabilize. Using the Kuder-Richardson 21 formula, “barrel rate” actually hit a reliability of .70 within a sample of 50 batted balls. As a point of comparison, HR/balls hit fair did so around 100 batted balls.

[Jim'sKid26]

56 minutes ago, Frobby said:

I’ve seen these kind of numbers, but I’m not really sure what “stabilize” really means in this context.   

Statistical stability:

"The consistency of results across samples, study designs, and analyses."

https://dictionary.apa.org/statistical-stability

 

[now]

This one stands out to me as the most surprising:

• 910 AB: AVG

In other words, nearly two full seasons to accurately predict the next...?