Pecota reality check - Fantasy Baseball Cafe 2015 Fantasy Baseball Cafe
100% Deposit Bonus for Cafe Members!

Return to Baseball Leftovers

Pecota reality check

Moderator: Baseball Moderators

Pecota reality check

Postby swyck » Sat Feb 26, 2011 12:20 pm

I enjoy reading Baseball Prospectus every year and like their insights into players and teams. I've gotten to some prospects ahead of the curve in the past based on these insights, e.g. Adrian Gonzalez, Brandon Phillips. I know Pecota projections get weird sometimes and it has to do with the whole range of outcomes that they use to hedge their bets but they're generally useful. However one projection this year just jumped out at me and made me shake my head.

Drew Storen is the Caps closer. They said as much in his write up and he doesn't really have any competition. He had 5 saves last year as a rookie and late season closer. They predict a reasonably good season in whip\era 1.27\3.50. Yet he's projected as finishing with ONE SAVE in 2011! That makes absolutely ZERO sense.

I understand that saves are irregular stats, based on opportunity, and projection systems don't like projecting them a lot. But how can you actually have a teams starting full time closer as having only one save. If you totally suck, but remain the closer for half a season on a terrible team you still wind up with 10-20 saves. They are not predicting Storen to suck, on the contrary they are expecting decent numbers from him.

This has to be a flaw with the way they model their projections, It can't be otherwise. I dont' care WHAT the reason is it is wrong, he can't be projected to get LESS saves than last year, he simply can't. They need to do some reality checks with their output since this one is a fail.

Yeah I know I'm overreacting lol but IMO this isn't just a diff between a 25 HR and a 35 HR projection, this is a model breakdown and it offends my stat geek sensibilities. :-b
swyck
Major League Manager
Major League Manager

User avatar
Cafe Ranker
Posts: 1524
(Past Year: 14)
Joined: 7 Feb 2005
Home Cafe: Baseball

Re: Pecota reality check

Postby Skin Blues » Sat Feb 26, 2011 1:54 pm

I'm not that familiar with Pecota, but I know that some sites that do projections that are highly stats based vs intuition based ignore saves altogether. They're not really tied to skill at all. Exhibit A: Fernando Rodney. The idea is that to begin the season, you'll have a pretty good idea who is each team's closer, or who is in the running, and can basically give all teams the same amount of saves to distribute amongst their pitchers. Even if they did offer a number of saves that Drew Storen would get, considering how early in the year they publish their magazine, would you really give it much credence?
Skin Blues
General Manager
General Manager

User avatar

Posts: 3082
(Past Year: 110)
Joined: 11 Apr 2010
Home Cafe: Baseball

Re: Pecota reality check

Postby swyck » Sat Feb 26, 2011 2:40 pm

Skin Blues wrote:I'm not that familiar with Pecota, but I know that some sites that do projections that are highly stats based vs intuition based ignore saves altogether. They're not really tied to skill at all. Exhibit A: Fernando Rodney. The idea is that to begin the season, you'll have a pretty good idea who is each team's closer, or who is in the running, and can basically give all teams the same amount of saves to distribute amongst their pitchers. Even if they did offer a number of saves that Drew Storen would get, considering how early in the year they publish their magazine, would you really give it much credence?


Yes, you are correct. Assuming a player is the full time closer, there is no way to accurately gauge the number of saves they will get. The best that can be done is do a rough saves\win calculation for that team based on recent history, and then whittle down a bit based on injury projections.

My point is that at the least they should have used that number, any number, other than one in their projection for Storen. Pecota doesn't ignore saves, though maybe they should if they're not going to make the effort. But as you point out, it doesn't matter anyway since I wouldn't be using that number for any kind of reference. I just like to see things neat and tidy.
swyck
Major League Manager
Major League Manager

User avatar
Cafe Ranker
Posts: 1524
(Past Year: 14)
Joined: 7 Feb 2005
Home Cafe: Baseball

Re: Pecota reality check

Postby spydog » Sat Feb 26, 2011 2:47 pm

Pecota is a good projection system, but they aren't really fantasy-centric. Up until a couple of years ago they didn't even project runs and RBIs. They have never done a great job at projecting playing time, either. I used them for a few years for fantasy and was fairly disappointed and didn't renew my membership in 2010.
spydog
Softball Supervisor
Softball Supervisor


Posts: 43
Joined: 12 Jan 2011
Home Cafe: Baseball

Re: Pecota reality check

Postby swyck » Sat Feb 26, 2011 3:05 pm

spydog wrote:Pecota is a good projection system, but they aren't really fantasy-centric. Up until a couple of years ago they didn't even project runs and RBIs. They have never done a great job at projecting playing time, either. I used them for a few years for fantasy and was fairly disappointed and didn't renew my membership in 2010.


I may not renew this year though I do love their baseball analysis and their articles. Not really made for fantasy but there are insights to be gleaned. I read them more from love of baseball then to help my fantasy leagues.

IMO they probably shouldn't project Runs, RBI, Wins or Saves based on how they do their projections. But if you're going to have projections do them correctly. It should probably be off some generic formula based on playing time, opportunity, and ability.
swyck
Major League Manager
Major League Manager

User avatar
Cafe Ranker
Posts: 1524
(Past Year: 14)
Joined: 7 Feb 2005
Home Cafe: Baseball

Re: Pecota reality check

Postby Skin Blues » Sat Feb 26, 2011 5:20 pm

The problem is that some people use their projections as-is without checking into players statuses. I imagine they don't want Joe Schmoe looking in their book seeing a bunch of saves for guys that lost a chance at the closer role well before the start of the season (Soriano is the main one, but I'm sure there will be at least a few more). That's pretty uniquely an issue for closers. Now like I said, I'm not familiar with Pecota, but I agree it's pretty dumb to not give any saves to Storen if they in fact give saves to other pitchers. Not really sure what the rationale behind that would be. Anway, for your purposes you probably don't want to use any saves projections that were published before mid-march anyway, so I wouldn't worry about it too much as long as you still like the other aspects of their product.
Skin Blues
General Manager
General Manager

User avatar

Posts: 3082
(Past Year: 110)
Joined: 11 Apr 2010
Home Cafe: Baseball

Re: Pecota reality check

Postby EWeaver » Sat Feb 26, 2011 7:49 pm

PECOTA is worthless (unless maybe you're Dallas McPherson's agent and you're trying to get a team to sign him any way you can).

One way to know for sure that it's worthless is this: if you could, with any significantly improved accuracy beyond educated guessing, predict the future, you wouldn't share that information with others.

I'm not saying PECOTA and Shandler and CHONE and who/whatever aren't fun or interesting to read, but c'mon. They have no greater ability in forecasting what's going to happen in 2011 than a reasonably informed baseball fan's intuition.
[
EWeaver
Softball Supervisor
Softball Supervisor


Posts: 57
Joined: 17 Feb 2006
Home Cafe: Baseball

Re: Pecota reality check

Postby Steve-o » Sun Feb 27, 2011 12:58 am

EWeaver wrote:I'm not saying PECOTA and Shandler and CHONE and who/whatever aren't fun or interesting to read, but c'mon. They have no greater ability in forecasting what's going to happen in 2011 than a reasonably informed baseball fan's intuition.


That's just ridiculous. The whole point of projection systems is to take out the bias of human's intuition. Pecota isn't the best projection system, and they haven't been for a while. But they certainly do a better job than a reasonably informed baseball fan.
In my fridge: Founders Kentucky Breakfast Stout (KBS), Central Waters Peruvian Morning, Bell's Batch 10,000, Dale's Pale Ale, Capital Brewery Amber, New Glarus Moon Man, and Guinness.
Steve-o
Major League Manager
Major League Manager

User avatar

Posts: 1879
Joined: 7 May 2004
Home Cafe: Baseball
Location: Finally back home. A sweet 15 mi from Miller Park.

Re: Pecota reality check

Postby thedude » Sun Feb 27, 2011 11:57 am

EWeaver wrote:One way to know for sure that it's worthless is this: if you could, with any significantly improved accuracy beyond educated guessing, predict the future, you wouldn't share that information with others.


A single person might guess better on certain individual players than the projection systems. But if you, "using educated guessing," made predictions for every single MLB player, you would see that PECOTA and the other projection systems would out-preform your guesses on the whole.
"I do not think baseball of today is any better than it was 30 years ago... I still think Radbourne is the greatest of the pitchers." John Sullivan 1914-Old athletes never change.
thedude
Hall of Fame Hero
Hall of Fame Hero

User avatar
CafeholicPick 3 Weekly Winner
Posts: 8413
(Past Year: 2)
Joined: 18 Dec 2004
Home Cafe: Baseball
Location: The Good Life

Re: Pecota reality check

Postby PlayingWithFire » Mon Feb 28, 2011 9:07 am

Here's a crazy one, Jay Bruce is projected for a higher WAR than Carlos Gonzalez this year...
Are you interested in joining a 28 teams dynasty league? If so, PM me.
PlayingWithFire
Hall of Fame Hero
Hall of Fame Hero

User avatar
Cafe WriterLucky Ladders Weekly Winner
Posts: 13262
Joined: 7 Apr 2005
Home Cafe: Baseball
Location: Manhattan, KS

Next

Return to Baseball Leftovers

Who is online

Users browsing this forum: No registered users and 8 guests

Forums Articles & Tips Sleepers Rankings Leagues


  • Fantasy Baseball
  • Article Submissions
  • Privacy Statement
  • Site Survey 
  • Contact