Off Season Projects
I have several projects to complete this off season to help fantasy owners for the 2012 season. Please let me know if there are any other ideas you would like me to examine over the off season. If I can’t get to the subject, others may be able look into the issue.
Fantasy “WAR”
Every player has factors that they can and can not control that factor into their stats. Pitchers can control the number of strikeouts and walks they allow, but they have little control over run support and defense. I want to get a player ranking system that shows the player’s true talent: contact, speed, and power. Then add in the factures they have no control over like the batters around them (more RBI and Run opportunites) and position in a lineup (more SB and Runs from the lead off spot).
The process will take a few weeks, but it should help an owner understand where the player gets their fantasy value and which surrounding changes affect the player.
Positional Rankings
Once the Fantasy “WAR”, or whatever it ends up being called, has been finalized, I will come up with a systematic way of ranking the players using talent, health, batting position and surrounding team talent. It will not be like what ZIPs does on Fangraphs where they guesstimate the final season stats. Instead, the values will constantly project the player’s talent using several factors.
Injury Information
Finding out how a player performs after an injury is important information. I am in the middle of examining pitchers coming back from TJS, so this information will be made public soon. Also, I will look into back injuries. They seem to really take a player off their game and seem to linger for a while. Let me know if there is any other specific injuries that you would like to see looked at.
Draft Rankings
I will be keeping my pre season overall rankings in a Google Doc with information and links on each player. It will be only the players I have examined in articles or otherwise. As I examine more and more players, the list will grow. I prefer to have a good understanding of a player than a vague notion from every player.
Reader Requests
Let me know what you want to see. It is a long off season, so I will be hurting for topics. Do you have keeper league questions? Ranking a few players? Draft strategy? Now is the time to let me know how I can help you answer any questions you have bouncing around. If I can’t find an answer, one of the other writers may certainly be able to find it.
Jeff, one of the authors of the fantasy baseball guide,The Process, writes for RotoGraphs, The Hardball Times, Rotowire, Baseball America, and BaseballHQ. He has been nominated for two SABR Analytics Research Award for Contemporary Analysis and won it in 2013 in tandem with Bill Petti. He has won four FSWA Awards including on for his Mining the News series. He's won Tout Wars three times, LABR twice, and got his first NFBC Main Event win in 2021. Follow him on Twitter @jeffwzimmerman.
RE: Fantasy WAR
Great idea. Something that I do for my league. To me the hardest part of implementing this idea publicly is establishing replacement level. The baseline will change depending on how many teams one’s league has. For a projection, I would also suggest having weights for statistics year-to-year correlation.
Once you have a means of determining Fantasy WAR, simply use the highest rated player on the waiver wire, or the average of the five to ten highest players, and set that as your replacement level.
This method works pretty decent. Also, I will rank the top X players and then the ones at the bottom are the replacement level. The replacement level setting will be the hardest.
Don’t do it. This will ruin those of us who try harder.
It’s true. The outcome of my league is becoming based more on luck and less on good decisions. Kinda like how fantasy football always was.
if I’m reading this right, I think your thinking is flawed on a couple levels.
– the biggest is that fantasy is not real life, and factors that we should ignore in figuring out real life value absolutely need to be considered when figuring out fantasy value. if I have a Padres pitcher and a Rockies pitcher with the exact same peripherals, the Padres pitcher will be more value in fantasy – but not real life – b/c of his park environment and majority of road parks he plays in.
– therefore, true talent is incredibly important in real life but somewhat less so in fantasy. a Red Sox scrub catcher will be more valuable than an exactly talented scrub Padres catcher.
– correlating your “true talent” with any one fantasy league will be extremely difficult or impossible. if Mariano Rivera has true talent of X, how does that affect his value in a 5×5 or 7×5 or 10×10 league? it clearly changes, but how and by how much?
– you really need a category based valuation system combined with a projection system which takes into account surrounding talent, park factors, etc. this stuff already exists
On your points 1. Agree. I think the only talent a pitcher has is K, BB and ability to keep the ball on the ground. Then the park (which I didn’t mention, but plan on including) and other team mates needs to be taken into account. 2. Agree, that is why I want to add in a team component. I would like to know how the player like AGon is going to change by moving to Boston from San Diego 3. I will have to work through each category on what % is talent and what is factors the player as no control over. 4. Agree
Would love to see how to think about relative values for each category assuming different deep leagues (10 teams vs 20). For instance, how many runs equal a win. whats more valuable a SB or a HR, etc.
Good idea. One problem we have with this work, is getting the baseline league data from 20 or so leagues with the same settings. I have added to my to do list and will get to it once I am further along in the valuation process.
Big fan of the idea of Fantasy WAR. As you are putting it together, please bear in mind that there is no one true rule set. Many many leagues don’t use standard 5×5 Roto scoring, for example. It would be awesome if we, the user, could dynamically select which categories to include.
A PITA I’m sure, but that’s what the off season is for right? 🙂
I think it may end up being partially user and partially automated. There is just too many variables. Also, what are the extra categories people are using? I will be able to look into each as I have time.
I mean things like OBP instead of AVG. Net Steals instead of raw SB. No way you could possibly think of everything, or include it even if you did.
If you’ve got a CBSSports custom fantasy league (or created one), there’s a page located at http://[your league url]/setup/custom-stats that lists a brazillion possible scoring categories as well as how many of their leagues actually USE each one. Might be handy…
I did not know that about CBS (I don’t play leagues there any more, it is not worth the frustration). I will get the data though.
May I suggest analyzing the effects of wrist injuries on hitters? Beyond just the power-sapping part (if that is indeed the primary side effect).
I looked at in detail after Sir Albert’s injury:
http://www.fangraphs.com/fantasy/index.php/albert-pujols-and-the-effects-of-the-wrist-injury/
My apologies – I look forward to reading it!
This might be a little far-fetched, but I have been interested lately in major swing mechanics and adjustments among batters. To be specific, instances like Bautista’s addition of a leg kick in 2009 (?) or Granderson’s re-tooled approach to lefties late last season. These under the radar adjustments have payed huge dividends for these players and I wonder if there are any high upside players this season who have made similar changes that would warrant 2012 investment?
I sort of have a method to find such players. I am looking to see if players begin pulling the ball more using batted ball data (Bautista, Tulo and Ganderson all did this). Let me see what I can find. I have added it to my to do list and will wait until I have all the of the 2011 data available.
how about an evaluation of the projection systems (ZIPS, Marcel, Fans, etc) listed on the fangraphs site? i realize a basic comparison has been done before. i personally was wondering if the fans are getting better at projecting player performance given the wider array of stats.
Tom Tango does a pretty good job of this comparison is his projection system challenge at the Book Blog. I would see what his final results are:
http://www.insidethebook.com/ee/index.php/site/comments/forecasters_challenge_current_results/
Awesome stuff. Would it be possible to also take a team’s opponents into account along with ballpark? On average, a pitcher in the NL West is going to get better results than his peripherals more often than a pitcher in the AL East due to the quality of opponents (though ballpark factor goes into that somewhat).
I started looking into this this and found very small differences. I may have to put together and publish the results.
one interesting way to rank players could be by category. even though a HR is massively more valuable than a SB in real life, they are the same in category leagues (so most leagues). so one can, instead of just searching out that 8 WAR player, can see that they’re set in HR and could really use some SB help and find those first basemen that have wheels instead of power. my conception of the presentation would be to give a player a score above or below the average for the major categories for their position. i dont know how feasible this is, but it’s just a suggestion.
this is brilliant. i look forward to seeing it come to fruition.
may be slightly off topic, and not sure where exactly to ask this question, but i keep seeing a link to the heat maps at the top of this page, so:
is there a way to sort the heat map data (and other pitch f/x data) by parameters such as runners on base, inning, pitching with lead/ from behind etc?
for example, i think it would pretty interesting to see Morrow’s heat maps with runners on base, vs no runners on base… and to see if his location/pitch selection/velocity/ etc changes significantly
if not, would this possibly be added to your list of Off Season Projects?
In my keeper league, we keep 10 and have three years of team control before a restricted free agent period (when players can be bid on with longer contracts or be added to the draft pool). We can also keep seven players on the bench at no charge, and their contract becomes $1 (for year one, $1 for year two, $6 for year 3) if you pull them from your bench to your lineup at any point in the season.
One thing I’ll be tasked with this offseason is weighing available free agent options’ value vs. pulling players from my bench. As projected now, I’m on the high end of the win curve for next year and hold an apparent advantage in understanding cost control vs. value.
Any of these players worth pulling up from the bench for next year? Any worth cutting to save one more bench spot?
Mike Trout
Des Jennings
Dominic Brown
Alex Rios
Wil Myers
Matt Moore
The broader question being — and this could go with WAR positional adjustments — but which positions are stacked and which are thin from a fantasy perspective. (The initial answer is obvious but to what extent are quality SS more valuable than, say, a quality RF)?