Odds of Eric Gagne's consecutive saves record being broken??

[Mays Hayes]

Can someone help me calculate approximate odds of Eric Gagne's record being broken based on league average save % and league average save opportunites/yr. What I'm looking for is the number of years for it to be likely that the record is broken.

I'm no stats guru, and the numbers I've come up with basically say that it is nearly impossible for the record to be broken based on a league average save %.


pretending that each team only uses one closer, is this equation even close to correct?
EDIT: No, it's not!! my calculations didn't consider the fact that the streak starts over after every failed attempt instead of every 85 opportunities.

Still if someone who knows what they're doing can give an idea of how likely it is that this record is broken over the next 50 or 100 years. etc. I'd be very interested... based on league ave save %'s and average save opportunites/yr.

 

[Baldkin]

Well, the opportunities/yr are irrelevant, as Gange did it over 2 seasons.

 

Basically this is a probability problem.

 

The odds of the first save, is 19/20 (assuming 95% is league average), so to figure out the odds of it happening 85 times in a row, you take (19/20)^85th, which comes out to be 0.0128. That's the odds of any given closer doing it. Multiply that by 30, and you get the odds of it happening in a given season.

 

My math may be off, its been awhile.

[Mays Hayes]

opp/yr is relevant in determining how many years pass before the record is broken... and your example like my original equation doesn't take in consideration the fact the the streak resets after any failed attempt not every 85 attempts. I know how to calculate the odds of it occuring in one attempt. What I can't figure out is how to calculate the odds of it occuring over a time frame of 50 yrs/100 yrs/ etc. (which is why opp/yr is relevent). league ave is typically a shade under 70% I believe.

[rluzinski]

This is very similar to when people try to estimate the odds for a particular hitting streak. You are right that you have to consider the opportunities. The easiest way is to just use a monte carlo approach. Really, that's the only way I've seen people approach it (not that there isn't a super secret probability equation to figure it out).

[dadofandrew]

Perhaps the biggest factor in this is how many times a guy had "easy" saves, tying run on deck, only need one out, or a 2-3 run lead with weak hitters coming up. In 57 games, that could be as high as 40-45, or as low as 25-30.

 

No way it's going to be done without a vast majority of "easy" saves. Bloops, seeing-eye grounders, errors, and a single bad pitch will too often get you in a tough situation.

[Mays Hayes]

I haven't been able to find any Monte Carlo studies done on Eric Gagne's consecutive saves record... I was only able to find simulations on oft discussed and publicized Joe DiMaggio's hit streak. These analysis indicate it is not extremely unusual that in the history of baseball that a player was able to hit in 56 games consecutively. However, given the high batting averages of the early 1900s late 1800s these studies don't necessarily give a great indicator of how difficult this record will be to break going forward. Dimaggio's record is commonly refered to as one of the "safest" records in the books.

Comparing the two accomplishments and their unlikelyhoods is difficult, the modern 1 inning save, the closer role, and the save statistic in general are relatively new to baseball. The closer role as we know it has really only been a fixture in baseball since everyone's favorite manager, Tony LaRussa, popularized the role with the help of Dennis Eckersley in the late 80s.

But consider this, I admit, rather parochial comparision.

A player that bats .400 during a 57 game stretch (4 AB/game) has a ~.87 or 87% chance of getting a hit in one game (1-(.6^4)
.87^57 = .000357 or .0357% chance of hitting in 57 consecutive games

These odds compare to a closer who saves .91 or 91% of his saves managing to close 85 consecutive games

.91^85 = .000330 or .0330% chance of saves 85 consecutive games

 

Now I don't have any stats for such events, but it doesn't take statistician to tell you the number of times a player bats .400 over 57 games happens A LOT more often than a closer averaging 91% saves closed over an 85 game span.

There are only 192 players that have 85 saves in their entire career. Baseball's best closers do not have a save conversion rate of 91% (Rivera .847, Eck .8125, Hoffman .868) Obviously, there just are not many chances for a closer to accomplish such an event, it takes two seasons for 1 closer/team to even log enough games to have a qualifying event. Compare that to the thousands of opportunties major league hitters have to bat .400 over a 57 games span.

Now, I realize that I haven't written any mathematical proofs here, but I think that you can see where I'm going with this. The more I think about, research this record, the more I think it is one of the most unbreakable, underrated, and unlikely feats in baseball history. Next to Cy Young's wins record I'm not sure there is another record more unlikely to be broken. Especially considering that the modern day "closer role" has no guarantee of being a mainstay in major league baseball for all of eternity. You never know when the next Tony LaRussa is going to come along and popularize a new and improved way of managing their pitchers to close out games. If I were a betting man, I'd bet everything I have, could beg or borrow for that none of us see this record fall or even approached in our lifetimes.

[logan82]

Are you looking at career save percentage? That may be somewhat misleading since relievers often start as middle relievers where they can easily pick up a blown save, but have virtually no chance of picking up a save.

[Mays Hayes]

yes, the percentages on Rivera, Eck, Hoffman are career numbers. I understand what you're saying, but it still doesn't change much in regards to the point I was trying to make on the consecutive saves record. Carrying a save percentage of over 90% over a span of 85 games does not happen very often.

 

A closer keeping his job for 85 save attempts is an accomplishment in itself.

 

Looking over the past few years of saves stats.....

 

In 2008, only 3 closers in the top 20 in saves converted 90% of their opportunites:

Soria, Lidge, and Rivera

 

Thus far in 2009 an impressive 11 of the top 21 closers are converting 90% of their opportunites.

But still, only Soria,Rivera, Hoffman are converting 90% since 2008, and that still doesn't get them to 85 opportunities.

 

To get over 85, Go back through 2007, Hoffman and Soria fall below 90%, leaving only Rivera closing at a 90% clip since the 07 season.

 

Other closers are very close to 90%, Papelbon, Nathan... but the point is, it's rare for a closer to be even 90% over 85 games...

 

And as I showed in my example even a 91% closer has only a .033% chance of closing 85 straight and if their is only a few guys every three years that can close 90% over a 85 game span... what kind of odds does that leave??

 

as the closing percentage drops, the chances of closing 85 straight drops drastically

90% = .013%

89% = .005%

88% = .002%

87% = .0007%

86% = .00027%

85% = .00010%

 

70% = .000000000006% (league average) !!