Save percentage incorrect?

[ChiefIlliniwek]

On a few occasions, I have noticed that a player's save percentage is not accurate based on his save and blown save stats. Has anyone else ever noticed this?

[blubbla]

I believe the game is adding the number of holds into save %.

Thus, the formula being used looks like this:

Saves / (Saves + Blown Saves + Holds) = Save Pct.

That probably accounts for the discrepancy you're seeing. Is it the correct way to figure it? Probably not.

[ChiefIlliniwek]

Hmm... that equation really doesn't make sense. If that is what's being used, it definitely should be corrected.

[blubbla]

The only reason I can figure that it's calculated that way is because a hold is awarded in a save situation, and thus, is being calculated as a missed save opportunity.

Steve-- if you see this, can you check into it? Thanks.

[Scott Vibert]

That's not the formula that OOTP uses.

The formula for Save Pct. is:

Saves / Save Opportunity


There can be more than one save opportunity a game, but only one save. Relief pitchers who enter the game in a save situation and leave with the save opportunity still intact are given a hold. But this hold doesn't help their save percentage.

Save Percentage is really only a relavant stat for closers.

For Middle Relievers Saves + Holds / Save Opportunity would be a better measure... but you must calculate it yourself.

I personally use Blown Saves more than the Save Percentage, because of the reasons stated, Save Percentage is only a relevant stat for your closer.

[blubbla]

Scott --

What I'm referring to is how the game appears to be calculating the stat, which is why saves and blown saves don't appear to jive.

That's why this might be something to look into.

[Scott Vibert]

What I posted is exactly how the game calculates the stat. It uses the Save Opportunity field.

Now it just happens that Save Opportunitys are equivilant to Saves + Blown Saves + Holds, but that is because the definition of a Save, Blown Save, and Hold requires that a Save Opportunity exist.

Save Percentage is simply a ratio of Saves to the number of Save Opportunities... which is why for Middle Relievers, who's save opportunities come in situations where they are awarded holds, and not saves, have poor save percentages. They aren't getting saves in their save opportunities. Hence the reason its not a valid metric for anyone other than a closer.

[blubbla]

I've seen the light....

Thanks, Scott.

[Chief Wahoo]

Reading this post made me think about something that I noticed while playing a game yesterday. I have a middle reliever who has 4 saves but only has had 3 opp. His save percentage was 133%. The only thing I could think of was that I brought him into a game that we were winning by like 7 runs but because he pitched the last 3 innings he recorded the save. Does the game not count this as a save opp?

[Scott Vibert]

Apparently not, I suspect the save op is triggered immediately when the pitcher enters the game... so if he eventually gets the save (3-inning flavor) the game hasn't charged him a Save Op. and as a result it registers as a bonus. I believe this needs to be fixed (although I do want to double check that MLB doesn't actually give someone a free save in this scenario).

[ChiefIlliniwek]

So wouldn't an accurate definition of Save % be...

Saves / (Save Opportunities - Holds)

[blubbla]

That's what I was trying to say, but apparently it is not.

[ChiefIlliniwek]

That's what I was trying to say, but apparently it is not.

A closer can get a hold too and it shouldn't count against his Save percentage.

[Jason Moyer]

From espn.com's stats glossary:

Save Percentage

Saves (SV) divided by Save Opportunities (OP).

Save opportunities

When a pitcher 1) enters the game with a lead of three or fewer runs and pitches at least one inning, 2) enters the game with the potential tying run on base, at bat, or on deck, or 3) pitches three or more innings with a lead and is credited with a save by the official scorer

In summary:

1. The game should be awarding save opportunities when a pitcher gets a 3-inning save.

2. The Sv/SvOpp formula used by the game is correct.

Jason

[ChiefIlliniwek]

I still don't understand this logic. How can a player be given a "Save opportunity" if he's taken out before the end of the game and therefore does not have the opportunity to get a save???

[Moocow]

I'm fairly sure OOTP4's definition of a save opportunity is incorrect. A pitcher with 10 saves, 10 holds, and 0 blown saves does NOT have a 50% save pct. He has a 100% save pct.

If you want an example of this, see one of ESPN's team pitching stats pages. The Chicago White Sox are a good example since they had a closer-by-committee system last year and a number of their pitchers recorded saves in addition to holds.

http://sports.espn.go.com/mlb/teampitchstats?team=chw

Here's a few examples:

Keith Foulke -- 11 Sv, 8 Hld, 3 BS, 14 SvOp (not 22 SvOp)
Damaso Marte -- 10 Sv, 14 Hld, 2 BS, 12 SvOp (not 26 SvOp)
Antonio Osuna -- 11 Sv, 9 Hld, 3 BS, 14 SvOp (not 23 SvOp)

And so on. I think ESPN's definition Jason gives is incomplete. A different site (http://www16.brinkster.com/bbstats/glossary.html) defines a save opportunity as saves plus blown saves. This makes much more sense than Holds being counted against your Save Pct.

I hope this can be fixed for OOTP5.

[HerbD]

The three inning save is a little different in that a pitcher must pitch effectively enough to warrant a save in the opinion of the official scorer. A three inning save is not automatic and is left up to human opinion which is a lot tougher to code in the game than the cut and dry save opportunity.

[kq76]

The way ootp calculates a save opportunity is faulty. Saves + Blown Saves + Holds does not account for the total Save Opportunities.

Let's think about it. If you really have an opportunity at a save, you'll either earn the save or you'll blow the save. If you're taken out of the game before you have the opportunity to earn the save, that cannot be counted as a save opportunity because you did not get the full opportunity!

For example, if Joe Schmoe enters the game in a save situation, records an out, and leaves the game never having relinquished the lead, he will get a hold. But did he get an opportunity at a save? No, he did not. So a save opportunity should not be added to his total!

What Saves + Blown Saves + Holds is closer to is a "Save Situation", which is completely different than a Save Opportunity. At first I thought you might include blown holds, but that would probably be the same as a blown save.

Now when Scott mentioned that he uses Blown Saves instead, I think he is onto something. The best reliever should probably have the lowest Blown Save %, calculated as Blown Saves / (Saves + Blown Saves + Holds). As long as there are holds included in Save %, it is not a good indicator of anything.

EDIT:

Looking at it closer, this is how I'm determining my relief pitcher award:

The player with the highest (# of Player i's Save Situations / # of Save Situations for Player with Most Save Situations)*(Player i's (1 - Blown Save %) / Highest % of (1 - Blown Save %) wins.

This weights equally highest # of save situations and lowest blown save %. Otherwise, a player with 1 save opportunity with no blown saves would win.

If anyone has any better ideas, I'd love to read them.

[Go Tribe]

http://mlb.mlb.com/NASApp/mlb/mlb/st...sortByStat=SVO

that list shows the SVO leaders from 2002 - unfortunately it doesn't show blown saves but you can see that SVO is closely tied to saves and NOT holds

here's the holds leaders

http://mlb.mlb.com/NASApp/mlb/mlb/st...sortByStat=HLD

the leader, J Romero, had 33 holds and zero SVO's. thus ESPN and OOTP are wrong

[Scott Vibert]

This has been fixed for OOTP5.