Pages

The Worst Three Weeks of My Life

11/19/2007
The NBA season is already one-eighth in the books. Three weeks down, 21 to go. Early surprises (Richard Jefferson, Manu Ginobili) and busts (Mehmet Okur, Kirk Hinrich) have gone from 'beginning of the season flukes' to 'sustainable trend'. The numbers are quickly becoming more and more real.

My fantasy squads are actually off to a very good start. Guys that I have weighted heavily like Kevin Garnett, Chris Paul, Ginobili and Andrei Kirilenko are performing well. I'm ahead of the pace compared to last season after three weeks. Right now, I'm projecting 35-42 wins out of 52 teams.

So why the gloomy title to this blog post? Well, the last three weeks have been marred by one of my squads and one player in particular. Dwight Howard. Owning Howard has been a nightmare.

Howard had been going very, very early in nearly all of my CBS drafts. You could bank on him going anywhere from 5th-8th, which was great news for me if I had a late pick because someone like Shawn Marion would usually slip through the cracks. But after a while I began to wonder if I was missing the boat on this guy. I had him about 17th-18th on my draft board, but there was zero chance he'd ever last that long. He looked good in the preseason, and there was some rumored video of his improved free-throw shooting on youtube. Indeed, if he could shoot 80% from the line, he'd be one of the top three fantasy players in the NBA. I'm sure most owners would be thrilled with just 72%.

Generally when I'm drafting I look at both my projected stars and the players average draft position. I usually try to avoid picking guys before their ADP. On one particular team, I had the 12th pick and took Dwyane Wade first. Usually when I had the last pick ( and I did in 15 out if 52 teams...I blame the astronomical odds on CBS Ron), I'd go with the Wade plus the best center of the three-headed Yao/Amare/Gasol monster. But all three of those guys were taken in the first 11 and there were no major surprises in the first round. So it came down to picking Paul or Howard. Yes, I'm still kicking myself.

I figured I'd either surround him with strong free throw shooters or just pick him purely for trade value. Ray Allen and Caron Butler usually made it to the 36th pick. But it didn't happen. I got Joe Johnson 36, Josh Howard 37.

So I'd have to try the trade route. Strangely, there were no logical trade partners. I tried offering him up every which way but Kevin Bacon, but to no avail. The guy with Amare was asking way too much, even with Amare's knee issues. The guy with Yao actually has a strong enough FT team to take on Howard and still be first in FTP, but he's my main competition for the team and I'd be helping him to a point where I probably couldn't catch him.

By all accounts, Howard is having a very good season. He's averaging 24.1 pts, 14.5 rebounds, shooting 58.6% from the field. That's great stuff. But then there's the free-throws...12.5 attempts per game at 63.0%. That's a team killer.

So after three weeks, I stand in fourth place in this league with zero chance of winning as long as Howard is on my squad. Dead Team Walking! It's killing me having a team that's already out of it.

I'm convinced that unless you have the perfect complementary squad, you just can't win in roto with Howard. Surveying my 52-team landscape, there are actually 4 teams that own him and are actually leading after 3 weeks, but I'm not projecting any of them to finish the season at the top. Most of these teams are winning in spite of him, generally flooded with 2-3 draft day steals in the pool of Ginobili, Jefferson, Chris Kaman, Josh Smith, Kevin Martin, Hedo Turkoglu. One team is dead last in FT and at or near the top in every other category. One team is actually 3rd in FTP. I don't expect that to last.

Contrast that with Yao Ming owners, who got Yao on average about 3-4 slots after Howard in the draft. 10 out of 52 teams with Yao are winning their respective leagues. The average Yao team is 8 points higher in the standings than the average Howard team.

I'm happy to report that after a lot of work, I was able to finally deal Dwight Howard, ending this three-week misery. I traded Howard along with Durant and Joe Johnson for Kobe Bryant, Marcus Camby and Brendan Haywood (who I will likely dump). I actually had Camby ranked ahead of Howard so essentially it's a decision of Joe Johnson and Kevin Durant for Kobe, which is a no brainer. The deal isn't a major steal, but I feel liberated. I'm projecting to gain 8 pts in the standings with this deal. Not quite in first yet, but relevant again.

Of course, now I have to deal with owning Marcus Camby now, but I'll take that headache over owning Dwight Howard and literally having no chance of winning.

PROJECTION UPDATE

It's probably way too early to look at how well those preseason projections are doing, but I have the numbers and it's relatively easy to do, so let's the take a look for the heck of it. It's a marathon, but let's take a look at how the competitors are doing after the first 3 miles.

I measure the projections based on their 'sum of squared errors' across the eight-categories and rank order them in two different ways. First I measure the average improvement versus the baseline of the previous year's per game averages (rookies excluded). Then I assign points for each projection across a pool of 232 players (216 initially as I set a minimum number of games cut-off at 4), the projection with the closest SSE gets 14 pts, the one with the worst projections gets 1 point.

Here are the early results.

IMPROVEMENT VERSUS BASELINE
1 COLTON INDEX 19.69%
2 CONSENSUS PROJECTIONS 18.15%
3 ROTOFREAK 15.56%
4 FANTASY SPORTS CENTRAL 13.68%
5 ESPN/TMR 13.46%
6 HOOPSKLYCE 12.58%
7 FANTASY HOOPS REALM 12.47%
8 BASKETBALL MONSTER 12.35%
9 NBA.COM 12.08%
10 ROTOWIRE/YAHOO 10.05%
11 ROTOWORLD 7.19%
12 CBS SPORTS 6.54%
13 2006 BASELINE 0.00%
14 FANTASY LOUNGE -17.80%

POINTS-BASED
1T COLTON INDEX 1892
1T CONSENSUS PROJECTIONS 1892
3 ESPN/TMR 1703
4 ROTOFREAK 1630
5 NBA.COM 1606
6 FANTASY SPORTS CENTRAL 1599
7 HOOPSKLYCE 1597
8 BASKETBALL MONSTER 1578
9 ROTOWIRE 1515
10 FANTASY HOOPS REALM 1490
11 ROTOWORLD 1459
12 2006 BASELINE 1444
13 CBS SPORTS 1421
14 FANTASY LOUNGE 1321

As expected, the combined consensus projections look like they will be a formidable (and perhaps the only formidable) projection opponent. And CBS is taking up its customary spot at the bottom of the food chain. At least it has some company down there with newcomer Fantasy Lounge (btw, don't mistakenly type in fantasylounge.com at work).

The boobie prize for worst single projection has to go to Fantasy Lounge for its projection of Chris Bosh. They projected 24.8 pts, 11.2 rebounds with an astonishing 57.7% FGP and 64.6% FTP. I had to go back and double check it to make sure that was in fact what they were projecting. Those projected shooting percentages are way off the reservation. You don't need to be a stats major to know that the SSE is going to be high when your actual totals are 16.2/6.9/40.2%/89.3%. I know Bosh has had a very strange start to his season, but I don't think we'll see anything approaching the numbers that Fantasy Lounge projected in this instance.

1 comments:

  1. Todd Lucas said...:

    Jim, thought that I'd check back on your blog for the first time in awhile. Like your Colton Index and hope to get some more college basketball analysis in the near future. Have to admit that I'm not a big NBA fan. I hope that doesn't count against me too badly.

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