Stabilization points for statistics - death to small sample sizes!

The Wolf · 07-08-2014, 06:40 PM

Just how many events are necessary for a given baseball statistic to overcome small sample sizes and actually become reliable?

From FanGraphs (original research by Russell "Pizza Cutter" Carleton over at Baseball Prospectus), here are the stabilization points:

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

Honorable_Pawn · 07-08-2014, 07:15 PM

I'm guessing this means when the stats become "significant". What margin of confidence error? 3% or 5%?

The Wolf · 07-08-2014, 07:32 PM

Quote:

Originally Posted by Honorable_Pawn

I'm guessing this means when the stats become "significant". What margin of confidence error? 3% or 5%?

Click the two links.

"These numbers represent the point at which each stat reaches a reliability of .70 or greater according to the relevant formula."

Honorable_Pawn · 07-08-2014, 07:43 PM

I'll have a closer look at the link. It looks like an interesting read. 70% seems low. I didn't know that a sample size could be statistically significant with such a low confidence %. Then again, stats is not my best subject.

Thanks, again for this.

The Wolf · 07-08-2014, 07:55 PM

Here's the original paper:

525,600 minutes: How do you measure a player in a year? | Statistically Speaking Archive

OmahaReynolds · 07-10-2014, 06:27 PM

It's a good read. Whenever I have a player who's off to a slow start and has a bad first month (or even a player on the team I root for in real life), I always look at BB, K and HR rates to get an idea as to whether or not they are just getting unlucky. That plus BABIP. If the player getting into his thirties, and it hitting .200 going into May, and his K% is way up from his career norms, I would think it's a safe bet that this player is getting into the decline phase.

07-08-2014, 07:15 PM	#2
Honorable_Pawn Hall Of Famer Join Date: Mar 2008 Location: Tampa Bay Posts: 6,407	I'm guessing this means when the stats become "significant". What margin of confidence error? 3% or 5%?

07-08-2014, 07:43 PM	#4
Honorable_Pawn Hall Of Famer Join Date: Mar 2008 Location: Tampa Bay Posts: 6,407	I'll have a closer look at the link. It looks like an interesting read. 70% seems low. I didn't know that a sample size could be statistically significant with such a low confidence %. Then again, stats is not my best subject. Thanks, again for this.

07-10-2014, 06:27 PM	#6
OmahaReynolds All Star Reserve Join Date: Aug 2010 Location: New York Posts: 653	It's a good read. Whenever I have a player who's off to a slow start and has a bad first month (or even a player on the team I root for in real life), I always look at BB, K and HR rates to get an idea as to whether or not they are just getting unlucky. That plus BABIP. If the player getting into his thirties, and it hitting .200 going into May, and his K% is way up from his career norms, I would think it's a safe bet that this player is getting into the decline phase.