Pages

Fantasy Hoops: 2007-08 Season Recap

4/18/2008
I've gotta admit...I am glad the 2007-08 NBA season is finally done. Had the fantasy season dragged on for another six weeks, I would've probably gone certifiably insane. The amount of injuries and random DNP's from high-caliber players is simply too much for one guy to take. In some ways, the 2008 season was just as bad as 2007 in terms on injuries, which is really saying something. Just don't get me started about Shawn Marion.

In general, it was a pretty disappointing season for my 52 fantasy teams, even though I did manage to win 20 of them and make a nice profit of about $12K (which equates to roughly 45 cents per hour, if you were wondering). But it's all for fun and science anyways. I guess I shouldn't complain. But about six weeks ago I was staring at somewhere between 30-32 wins. But it's safe to say if you had teams built around Yao Ming, Dwyane Wade, Shawn Marion (don't get me started), Chris Kaman, Gerald Wallace, Caron Butler, Manu Ginobili, Mike Miller, Kevin Martin and Maurice Williams, you probably didn't fare too well down the stretch. Thankfully, I had enough of a cushion in some leagues to hold on..many cases just barely (as in 1 or 2 field-goal makes barely). Thank goodness for Chris Paul, Marcus Camby and Lamar Odom, the only guys I had big exposure to that actually played well down the stretch.

THE PROJECTIONS COMPETITION - FINAL RESULTS

If only my fantasy teams had been as successful as my preseason stat projections (maybe I should've listened to my own numbers and drafted Mike Dunleavy, but I just couldn't pull the trigger on him). The stat projection competition really wasn't much of a competition at all, I had the best projections so far. In fact, I finished ahead of the combined consensus projections, so you could say that I outperformed all of the other projections combined. If you received the projections before the season, I hope it helped lead you to the title.

As I've done in the past, I look at the results two ways. The first measures the improvement in the projections across category (measured by sum of squared-error) using the 2006-07 numbers as a baseline (the crudest projection set you could come up with). The second ranking looks at each player and gives 14 points for the closest projection down to 1 point for the worst. Click here if you want a detailed run-down of each of the contenders.

IMPROVEMENT VERSUS BASELINE
1 COLTON INDEX (free) 26.63%
2 CONSENSUS PROJECTIONS 22.51%
3 ROTOFREAK (free) 15.86%
4 NBA.COM (free) 13.54%
5 HOOPSKLYCE ($16) 13.14%
6 FANTASY SPORTS CENTRAL (free) 12.05%
7 FANTASY HOOPS REALM ($14.95) 11.89%
8 ESPN/TALENTED MR ROTO (free) 11.80%
9 BASKETBALL MONSTER ($14.99) 9.75%
10 ROTOWORLD ($14.99) 9.04%
11 ROTOWIRE/YAHOO ($14.99) 6.50%
12 CBS SPORTS (free) 4.26%
13 2006-07 BASELINE 0.00%
14 FANTASY LOUNGE ($7.99) -4.45%

Fantasy Lounge gets the boobie prize for coming out worse than the previous season baseline. Previously, this spot was reserved for CBS but they managed to get above the Mendoza line. Just think about how hard this is to do. A monkey would use the previous season as a starting point and adjust from there. The adjustments that a so-called fantasy expert would make to this baseline in a set of projections should add value, not destroy value.

POINTS-BASED
1 COLTON INDEX 2107
2 CONSENSUS PROJECTIONS 2097
3 ROTOFREAK 1786
4 ESPN/TMR 1786
5 BASKETBALL MONSTER 1717
6 FANTASY SPORTS CENTRAL 1674
7 HOOPSKLYCE 1670
8 NBA.COM 1602
9 ROTOWORLD 1582
10 2006-07 BASELINE 1558
11 ROTOWIRE 1549
12 FANTASY HOOPS REALM 1539
11 CBS SPORTS 1528
14 FANTASY LOUNGE 1510

I'm a little surprised to have finished ahead of the consensus picks considering the consensus is going to be consistently solid across the board. In other words, at worst the Consensus is going to finish in the middle of the pack for any one pick and will more likely be near the top, while my projections could be last (or near last, at least ahead of the Lounge) for any given player. I'm also surprised to see four projections below the 2006-07 baseline.

To dig a little deeper, let's look how each performed in measuring the rookies, which in my opinion is the toughest thing to do. This is where the proverbial bread is buttered. Fantasy Sports Central took the cop-out route and opted not to make projections on rookies at all (what's the point in that?). So here's the top 12...obviously you can't measure versus the 2006-07 baseline so I'll show the results Nascar style and display the results in terms of how they were behind the leader. The results are based on 12 rookies that were common to each projection set (with the exception, again, of Fantasy Sports Central...I just can't get over the fact that they'd choose to leave the rookies out).

ROOKIES - SUM OF SQUARE ERRORS
1 COLTON INDEX
2 CONSENSUS PROJECTIONS 5.04% worse
3 BASKETBALL MONSTER 12.42%
4 ROTOFREAK 19.01%
5 ESPN/TALENTED MR ROTO 28.55%
6 FANTASY LOUNGE 36.11%

7 ROTOWIRE/YAHOO 40.06%
8 FANTASY HOOPS REALM 46.15%
9 HOOPSKLYCE 46.88%
10 ROTOWORLD 54.03%
11 CBS SPORTS 106.17%
8 NBA.COM 178.94%

The points-based method pretty much shows the same thing so I won't bore you with the details.

One area I spent a lot of time on this year was trying to project games played. The results prove what I already knew...games played is nearly impossible to predict. I was able to be a little more precise than the competition, but was still off by about 10 games on average per player. Here's the sum of squared error results for the 10 projections that bothered trying to project games relative to the 2006-07 baseline:

GAMES PLAYED
1 COLTON INDEX 21.67% improvement vs 2006-07
2 CONSENSUS PROJECTIONS 15.14%
3 FANTASY SPORTS CENTRAL 14.90%
4 ROTOWIRE/YAHOO 14.45%
5 NBA.COM 13.88%
6 CBS SPORTSLINE 13.23%
7 FANTASY HOOPS REALM 12.90%
8 BASKETBALL MONSTER 12.58%
9 HOOPSKLYCE 12.42%
10 ROTOWORLD 8.62%
11 2006-07 BASELINE 0.00%

Hopefully, you find this information interesting and useful in deciding which projection to buy or not to buy next season.

If I get up enough courage, I may take a closer look at some of my draft picks and trades, which promises to be a rather depressing exercise. I can only cringe at the thought of finding out how many times I picked Dwyane Wade over Chris Paul or Elton Brand over Hedo Turkoglu.

0 comments:

 
Wegoblogger #31 © 2011 | Designed by Bingo Cash, in collaboration with Modern Warfare 3, VPS Hosting and Compare Web Hosting