Fantasy Baseball Advice

As some of you know, we developed our own Player Rater methodology called Point Shares about a year ago. Since then, it’s been one big poontrain….zooming past Statgeek Station. Perhaps it’s because my hat and eyewear aren’t goofy enough?

Anyway, we’ve done some informal comparisons in the past w/ ESPN but – after having a prolonged, dorktastic debate on another blog – I decided to take it one step further.  I created a test where I pitted our Point Shares against two other player raters:  ESPN and RotoTimes.

The test went like this:  Create 10 random 10 team leagues and calculate the standings using the real stats.  Then do the same using the rating points for each of the services (so if you add up the ESPN HR player rater points for each team, the one with the most points gets a 10 in HR).  The team point totals calculated by the player raters is compared to the point totals from using the real stats and we added up all the differences (so if the real points total was 70 and the player rater projected 68, it’s worth 2 points).  The lower the point total the better.

Before we go into the results, it’s worth noting that this test puts our Point Shares at a disadvantage.  Why?  Our methodology factors in position and projected team standings so that you can use it to directly estimate the impact of a trade or draft pick.  We adjust our ratings by crediting points against the average found at the player’s position instead of the average hitter.  All stats equal, we know that a catcher will be much more valuable than a 1st baseman (why else would people draft Russ Martin and V-Mart in the 3rd round last year?).  Geovany Soto and Justin Morneau both hit 23 HRs.  While RotoTimes and ESPN ignore position and credit each with the same number of points, we have a 0.5 difference.  This is saying that – assuming you fill the rest of the rosters up with average players – owning Soto will give you a 0.5 advantage in HR points vs. owning Morneau in a 10-team league.

But a test where you just add up rosters takes this out of the equation.  You could conceivably be perfect just by creating a ‘rating point’ completely proportional to the real total.  So 20 vs. 30 HRs could be worth .2 and .3 points or 2 and 3 points, etc.  Will those numbers help determine the value of 20 vs. 30 HRs?  Not really.

Okay, with that said, the results of the test are as follows (if you want to see the spreadsheet, click here – warning: it’s messy).  Out of 100 teams, the total points off by the Player Raters was:

RotoTimes – 112
Razzball Point Shares – 143
ESPN – 152

Our Point Shares were the most accurate for 3 of the 10 leagues while RotoTimes claimed at least a share of 1st in the other 7 (ESPN tied them 3 times).

At a category level, I performed correlation tests between the ‘real’ standings and those of the player raters.  100% would be perfect.

As you can see, Point Shares held up pretty well to ESPN and RotoTimes despite taking on the extra burden of factoring in position and tying it to expected point gains/losses in the standings.  The test did shine light on some improvements we can make on pitcher counting stats (W, SV, K) – note the higher correlation we have on ERA/WHIP because we can directly tie it to IP.

Final summary:

1) RotoTimes is the most accurate of the three.  Combined with the facts that you can both customize the rankings based on league size, categories, roster size, etc. and they provide $ estimates, we’d say this is the best pure ranking tool.

2) Point Shares have proven to be in the same league as these other player raters.  Given the extra utility of Point Shares because it directly estimates the impact on team points of player moves, we’d say our Point Shares are the most useful of the three.

3) ESPN, while not the ‘testwide leader’, did pretty well – especially in hitting stats.  Their biggest pure weakness is ERA/WHIP as it would appear that they don’t factor in IP like Razzball (and, we assume, RotoTimes).  But we’d say their biggest total weakness is that their player points mean ABSOLUTELY NOTHING.  They are arbitrary numbers that could be used to rank players but not to estimate their value in the standings or worth at the draft table.  They do a good job, though, at keeping it updated throughout the year…

Not sure how many of you knuckleheads explore this site, but in the top header under ‘08 Rankings, which has the 2008 Fantasy Baseball Rankings, our Fantasy Baseball 10 Team Player Rater and our Fantasy Baseball 12 Team Player Rater. So now instead of downloading a clumsy Excel spreadsheet, you now have it all at your fingertips right onsite. Under ‘08 Rankings, you also have the Best Fantasy Values for 10 team leagues and Best Fantasy Values for 12 team leagues. So what gives? Did Christmas/Chanukah/You Name It come early? You better believe it. No, the pleasure is all mine. No, mine. Mine! If you need an explanation on what the player raters are or how to use them, look here:  Fantasy Baseball Player Rater — that will send you deeper into the player rater rabbit hole and you’ll come out a week later much wiser. The comments are off on this post. Scroll down to the next post to drop some praise on Rudy, this is all his doing. (Or scroll up if you’re finding this after today — the, um, day this was posted.)

Back in the pre-season, we launched a new player rater methodology called Point Shares to estimate fantasy baseball player value.   There aren’t a lot of Player Raters to be found other than ESPN (Y! and CBSSports.com don’t have ones) but we feel ours is better because it factors in variables like a player’s position (e.g., Hanley Ramirez’s 33 HR is worth more than David Wright’s) and the point totals actually represent something.

A ‘point share’ is the estimated impact on a team’s points by substituting a player for the average drafted player at his position on a team filled with average players.  So in a 10 team league, this team would otherwise earn 5.5 points per category (55 points).  Substituting Cliff Lee for Zack Greinke on an average roster would mean an average of 6.6 points (6.40 for Cliff Lee, -0.24 for Greinke).

Our 2008 Player Rater that is based on a 10 team MLB league with 5×5 scoring.  Only the top 50 are displayed below the post.

We will use this as the foundation for future articles but here are answers to some anticipated reactions:

1) How could pitchers be in the top 4 slots?

It might be surprising to see starting pitchers in the first 4 slots.  This isn’t a flaw in the ratings- it’s a flaw in your thinking (sorry).  The best pitchers in a season generally dominate 4 categories (W, K, ERA, WHIP).  While they don’t contribute in Saves, there is little opportunity cost to this since saves are concentrated in 30-40 players and there are 90 pitching roster spots.  For hitters, it’s rare for a player to be excellent across more than 3 categories and there’s a larger opportunity cost for the categories they are not great (i.e., OFs like Ichiro and Taveras hurt HR/RBI).  Only 6 hitters in 2008 ranked above average across all 5 categories – Pujols, Wright, Berkman, Holliday, A-Rod, and Braun.  Hanley Ramirez had a great year in Runs (125), HR (33), and SB (35) but was mildly valuable in AVG (.301 or .006 less than Ryan Theriot) and negative in RBI (67 or tied for 6th among SS).  Converting Roy Halladay’s stats to an OF (W=R, HR=SO, RBI=WHIP, SB=SV, AVG=ERA) would net 119/34/121/10/.335.  Sabathia, Lincecum, and Cliff Lee weren’t far off this.  The reason why it’s usually not advisable to draft a pitcher in the 1st round is the unpredictability of who’ll be the top pitchers (none of the top 4 were top 50 picks based on MockDraftCentral’s Average Draft Position) vs. their true value.  Here’s a more thorough explanation of a pitcher’s fantasy value.

2) Why are there are only 110 players that have positive value when there are 230 open roster spots?

In Point Shares, 0.0 represents an average player at their position.  So it is expected that about 1/2 the eligible players are above average, 1/2 are below.  A negative score doesn’t mean that a player isn’t worth having on your roster – it just means that he’s below average.  If the system was built with 0.0 representing a roster-worthy player, the points would then represent the value of a player on a team in last place.  This overestimates the value of a player when measuring the impact of a player swap (e.g., Lee for Greinke) for every instance except when a team actually did finish last.  The best approximation of player value is starting from the middle.

3) What is the highest/lowest score possible?

Theoretically, the best possible score should be 4.5 pts in each category (moving someone from 5.5 to 10) or 22.5 for a 10 team league.  Any other league size, just divide the team total by 2 and subtract 0.5 (12 team = 5.5).  A simpler way of looking at it would be to consider how one would calculate the average value per team of a player with 50 SBs.  For the team that’s 1st in SBs, his value is 0.  For the 2nd team, it could be worth +1.  Take this to its logical conclusion and you get (0+1+2+3+4+5+6+7+8+9)/10.  That’s 45/10 = 4.5.

While no hitting scores fall above 4.5 or below 4.5 (closest is Willy Taveras’s 4.3 in SB), the Point Shares system is just a model so it is conceivable that an extraordinary performance would net more than 4.5.  The only cases in 2008 were extraordinary bad ERAs and WHIPs.   (You’d think K-Rod’s 62 SV but that only netted +3.9).

Back in the preseason, we launched a new method for rating players called Point Shares. The objective was to create a methodology that converts player’s projected or real stats into a metric that would allow for easy comparisons across players. The metric we settled on is deceptively powerful – Roto points.

Assuming you have a completely average team (55 points in a 5×5 10 team league), each player’s point shares reflect how many points they would add/subtract from your team’s totals if they replaced the average stat projections at that position. For example, if you swapped out an average team’s 1B (Mark Teixeira, David Ortiz) with Lance Berkman, they would increase from 55 points to an estimated 67.5 points. Point Shares also allow for basic trade comparisons (add the points of each side) and assessing your team vs. others (just add up the points and add to 55).

Seeing that we’re 2 months into the season, we thought it was a good time to update our Point Shares rankings – in this case, it is based solely on actual stats through June 8, 2008. So this is more comparable to the ESPN Player Rater vs. ESPN’s player rankings or Yahoo’s Big Board. Except better. (Not saying there aren’t flaws in ours but, c’mon, Ichiro is #17 among all players with a .290/43/3/17/26? At OF?)

June 8th – Point Shares for 5×5 10 team league and for Fantasy Razzball (play for worst team)

(Click on images to enlarge)

Pre-Season – Projected Point Shares for 5×5 10 team and 12 team leagues

Pre-Season – Projected Fantasy Razzball Point Shares

During the offseason, I wrote several posts criticizing the ESPN Player Rater and went as far as create my own to critique it.

Below are the links to these posts. Interesting reading if you’re a fantasy baseball or stat junkie. Much less interesting than smack if you are just a junkie.

Part 1: How Valid Is The ESPN Player Rater?

Part 2: Are Fantasy Baseball Pitchers Correctly Valued By Player Raters?

Part 3: How Do You Value Fantasy Baseball Hitters?

The whole journey to prove/disprove the ESPN Player Rater eventually led me to the Point Shares concept that is the foundation of my 2008 drafting and player valuation.

Anyway, I checked in on the ESPN Player Rater in 2008 and it looks like they listened to good ol’ Rudy. Some of the improvement I’m seeing:

• Negative values (particularly in AVG, ERA, WHIP) – Where 2007’s version would have a floor at 0, the current rater penalizes very bad performance. CC Sabathia currently sports -3.31 points in ERA and -2.64 in WHIP.
• No arbitrary cap at 5 points per category – I’m seeing some category leaders at 3 and Michael Bourn at a 7.64 for SB.

It’s too early to do a full assessment of the methodology. I figure I’ll do a Point Shares update at the All-Star Break and compare it then. But given the shit I gave ESPN during the offseason, it’s only fair that I give them credit for coming to their senses (and listening to me).

Has anyone in Razzball Nation been using ESPN for their fantasy league? If so, post your feedback! Grey and I went cold turkey on ESPN FLB after last year’s debacle (first two weeks of stats were fried – I had to calculate the data offline and make year-end adjustments). But I thought their new platform was pretty good. Interesting to hear what the rest of y’all think.

Oh, and just in case the worldwide leader is listening, add smart people to Baseball Tonight. Replace Eric ‘The Emmitt Smith of ESPN Baseball” Young and John Kruk with guys from ESPN.com like Keith Law and Rob Neyer. If you must pander to the masses, go with some tail like Erin Andrews.