USCL Predictions for Week 6

Predictions for week 6 (again, no looking at the lineups, using only the Famous Original Power Rankings).

NJK	1.5		CAR	2.5	CAR wins
MIA	2		QUE	2	Tie Match
DAL	2		PHI	2	Tie Match
BAL	2.5		TEN	1.5	BAL wins
BOS	2		SFM	2	Tie Match
NYK	1.5		SEA	2.5	SEA wins

Directly outputted from the spreadsheet itself!

Famous Original Power Rankings, USCL Week 5

The results are in for the Famous Original Power Rankings, after the United States Chess League, Week 5.

This week, the power rankings did not change too much, only a few transpositions. Carolina and Dallas swapped places, as did Baltimore and San Francisco, as well as New York and New Jersey. You may ask how, after San Francisco's sweep of Tennessee, it was possible for San Francisco to drop a place in the power rankings. It is because of the strength of schedule factor -- Tennessee makes San Francisco's schedule easier. Conversely, Baltimore's tie with first place Boston, pushed them up in the rankings.

Enough said, let's get to the numbers.

1 -	Boston	1.000
2 -	Seattle	0.961
3 -	Carolina	0.934
4 -	Dallas	0.930
5 -	Baltimore	0.908
6 -	San Francisco	0.906
7 -	Philadelphia	0.839
8 -	Queens	0.778
9 -	New Jersey	0.748
10 -	New York	0.741
11 -	Miami	0.710
12 -	Tennessee	0.635

The interesting thing now, is that the other teams have bunched closer to Boston. Indeed, they are bunched closer to each other. In fact, only 7/1000 of a point separate New Jersey and NEw York, only 4/1000 of a point separate Carolina from Dallas, and a measly 2/1000 of a point separate Baltimore and San Francisco.

Next week, it is all Wednesday night games, so tune in for all the action as the games are played over the Internet Chess Club, and the league itself is sponsored by PokerStars.com.

USCL Predictions Week 5

The Famous Original Power Rankings have been used to generate this week's predictions (again, all done by mathematical model, with no human jiggering of the results).

Dallas over Miami 2.5 - 1.5
New Jersey ties New York 2.0 - 2.0
Boston over Baltimore 2.5 - 1.5
Queens ties Philadelphia 2.0 - 2.0
San Francisco over Tennessee 2.5 - 1.5
Carolina ties Seattle 2.0 - 2.0

In the next week or two, I will try to add a better parameter estimation model (rather than doing it by hand, like I am now).

Here's the USCL News Scoring Scale (quoted from their blog).

For each correct result picked you receive 2 points. You also receive the following bonuses for picking the exact final score as follows:

Each exact score of 2.5 - 1.5 receives 1 bonus point.
Each exact score of 2 - 2 or 3 - 1 receives 2 bonus points.
Each exact score of 3.5 - 0.5 or 4 - 0 receives 3 bonus points.


UPDATE TWO:
Predicted: DAL 2.5 - MIA 1.5 Actual: DAL 2.5 - MIA 1.5 CORRECT (+3)
Predicted: NJK 2.0 - NYK 2.0 Actual: NJK 2.5 - NYK 1.5 WRONG
Predicted: BOS 2.5 - BAL 1.5 Actual: BOS 2.0 - BAL 2.0 WRONG
Predicted: QUE 2.0 - PHI 2.0 Actual: QUE 1.5 - PHI 2.5 WRONG
Predicted: SFM 2.5 - TEN 1.5 Actual: SFM 4.0 - TEN 0.0 CORRECT DIRECTION (+2)
Predicted: CAR 2.0 - SEA 2.0 Actual: CAR 2.0 - SEA 2.0 CORRECT (+4)

That's +9 on the USCL News Scoring Scale. How did everyone else do?

USCL Power Rankings After Week 4

Well, in a week, the power rankings have changed quite a bit. Boston remains at the top, but Seattle is very close to them, after crushing San Franciso 3 - 1. In their first match loss since the 2005 semi-finals, San Francisco drops to 5th place.

In the battle of the New York City metropolitan region teams, Queens win over New Jersey, coupled with New York's tie with Boston, put Queens ahead of New York and both are ahead of New Jersey. Next week's match up between New York and New Jersey will alter that balance.

Also, in the battle of the basement, Miami beat Tennessee, and thus, has the pleasure of moving up a notch, and the Tempo go back down to the cellar.

So, here are the power rankings after week 4.

1 -	Boston	1.000
2 -	Seattle	0.950
3 -	Dallas	0.902
4 -	Carolina	0.881
5 -	San Francisco	0.867
6 -	Baltimore	0.841
7 -	Philadelphia	0.792
8 -	Queens	0.774
9 -	New York	0.743
10 -	New Jersey	0.696
11 -	Miami	0.684
12 -	Tennessee	0.631

Until next week! Be on the lookout for predictions... Coming soon!

UPDATE: The Boylston Chess Club bloggers have published their rankings...

1 - Boston
2 - Seattle
3 - Dallas
4 - Carolina
5 - Baltimore
6 - San Francisco
7 - Queens
8 - Philadelphia
9 - New Jersey
10 - New York
11 - Miami
12 - Tennessee

Their power rankings are very similar to our Famous Original Ray's Pizza Power Rankings. Just transpositions at 5-6, 7-8, and 9-10.

New Rankings

Oops. I goofed. After checking the math in the ranking system I published a few days ago, I realized that I had a math error, which overvalued the records of individual teams, and undervalued their opposition. So, here is the revised Power Rankings after Week 3.

1 -	Boston	1.000
2 -	Dallas	0.855
3 -	San Francisco	0.844
4 -	Seattle	0.838
5 -	Carolina	0.777
6 -	Philadelphia	0.770
7 -	Baltimore	0.707
8 -	Queens	0.703
9 -	New Jersey	0.661
10 -	Tennessee	0.643
11 -	New York	0.628
12 -	Miami	0.586

Not a lot different than before, but now correct!

The Tart Green Line

The Boylston Chess Club blog has claimed to initiate legal proceedings against the Bionic Lime Power Rankings, because in an earlier post, I claimed that my USCL power rankings were the only strictly objective ratings. I was wrong. I admit it. The BCC's power rankings are also said to be objective, and I have no reason to doubt them. Here is their article on Week 3 rankings.

The subjective power rankings I was referring to were the "official" USCL power rankings. It isn't on the USCL site, but they claim to be official. Hmm..

Anyway, lets compare the three ranking systems...

Rank	Official	Boylston	Lime
1 -	Boston	Boston	Boston
2 -	San Francisco	Seattle	Dallas
3 -	Dallas	Dallas	Seattle
4 -	Seattle	San Francisco	San Francisco
5 -	Baltimore	Carolina	Carolina
6 -	Philadelphia	Philadelphia	Philadelphia
7 -	Carolina	Baltimore	Baltimore
8 -	New Jersey	New Jersey	New Jersey
9 -	New York	Queens	Queens
10 -	Queens	New York	Tennessee
11 -	Miami	Tennessee	New York
12 -	Tennessee	Miami	Miami

The computer rankings are rather similar, the only difference is the transpositions of Dallas and Seattle and New York and Tennessee. The human rankings put New York (based on reputation, presumably, and rightly so I suppose) higher than the computer models. And, the human puts San Francisco second, again based presumably on reputation. Everybody agrees, though... Philadelphia 6th, New Jersey 8th.

This week's Boston versus New York may have a dramatic effect on the rankings, if New York draws or even wins. But, if Boston wins, it should not change the ranking too much at all.

We shall see tomorrow night!

UPDATE:

There is another set of subjective (i.e., human generated) rankings, which have been posted here. To summarize them:

1. Boston
2. Seattle
3. Dallas
4. San Francisco
5. Carolina
6. Philadephia
7. New Jersey
8. Baltimore
9. New York
10. Queens
11. Tennessee
12. Miami

Not too different, except that Baltimore is lower than the other systems. And, not surprisingly, New York is high, like the other human system.

USCL Predictions Week 4

Using the power rankings I generated, and developing a predictive model that takes into account the difference between the two power ratings, applies it to a lookup table, the model has generated the following predictions for Week 4 of the United States Chess League.

UPDATE: The goal of these predictions is to do this without knowing the lineups -- the power rankings take into account team performance only (not individuals) and so, therefore, does the model. It is risky, for sure, but it is a one parameter model, and I'm curious more than anything on how well it can do.

So, here they are for week 4.

New York Knights vs. Boston Blitz - Boston wins 3-1
Queens Pioneers vs New Jersey Knockouts - Tied Match 2-2
Philadelphia Inventors vs Baltimore Kingfishers - Tied Match 2-2
Tennessee Tempo vs Miami Sharks - Tied Match 2-2
Dallas Destiny vs Carolina Cobras - Tied Match 2-2
Seattle Sluggers vs San Francisco Mechanics - Tied Match 2-2

What a dull set of predictions!

Ok, yes, it is dull, but look at the power rankings...

NY (11) vs. BOS (1) -- wide disparity
QUE (9) vs. NJK (8) -- next to each other in the list
PHI (6) vs. BAL (7) -- also next to each other
TEN (10) vs. MIA (12) -- only one team separates
DAL (2) vs. CAR (5) -- two teams separate
SEA (3) vs. SFM (4) -- next to each other

So, the only one that seems like it is might be a decisive prediction is Dallas - Carolina, and indeed, a slight tweak in the model would make it so that it would predict 2.5 - 1.5 for Dallas, but there is no human intervention in the model output.

Lots of teams close to one another, and it should be an interesting week!

USCL Power Rankings, Week 3

So, I've decided to follow the crowd, and create power rankings for the United States Chess League. There are others that have power rankings, but my rankings have no subjective elements -- they are completely objective.

The formula for computing them is a combination of Match Points and Game Points of a team, and to a lesser extent the MPs and GPs of the team's opponents (strength of schedule), and to an even lesser extent, the MPs and GPs of the team's opponents' opponents. These are combined to produce a raw score. The best team is given a final indexed score of 1.000 and other teams are given a lower final score, depending on the ratio of their calculated raw score to the best team's raw score.

So, without further ado, here are my power rankings for the teams in the United States Chess League after the conclusion of Week 3.

1 -	Boston Blitz	1.000
2 -	Dallas Destiny	0.855
3 -	Seattle Sluggers	0.823
4 -	San Francisco Mechanics	0.808
5 -	Carolina Cobras	0.753
6 -	Philadelphia Inventors	0.742
7 -	Baltimore Kingfishers	0.670
8 -	New Jersey Knockouts	0.636
9 -	Queens Pioneers	0.589
10 -	Tennessee Tempo	0.539
11 -	New York Knights	0.535
12 -	Miami Sharks	0.472

Note that there is a wide gap between the first place Boston Blitz (1.000) and the second place Dallas Destiny (0.855), but a very narrow gap between the tenth place Tennessee Tempo (0.539) and the eleventh place New York Knights (0.535).

Unsurprisingly, the Blitz remain in first place with their convincing win over Philly. The Dallas Destiny moves up to second place, rising in the rankings. The New Jersey Knockouts remain close to eighth place for the third week in a row. And, the preseason favorite New York Knights have escaped the basement, which now is occupied by the Miami Sharks.

I've retrofitted power rankings for week 2. I've also done week 1, but there is too little data to make accurate predictions.

Week 2

1 -	Boston Blitz	1.000
2 -	San Francisco Mechanics	0.944
3 -	Dallas Destiny	0.905
4 -	Philadelphia Inventors	0.864
5 -	Seattle Sluggers	0.792
6 -	Carolina Cobras	0.775
7 -	Baltimore Kingfishers	0.666
8 -	New Jersey Knockouts	0.666
9 -	Queens Pioneers	0.647
10 -	Tennessee Tempo	0.614
11 -	Miami Sharks	0.497
12 -	New York Knights	0.468

Week 1

1 - 3	San Francisco Mechanics	1.000
1 - 3	Dallas Destiny	1.000
1 - 3	Baltimore Kingfishers	1.000
4 - 5	Boston Blitz	0.967
4 - 5	Philadelphia Inventors	0.967
6 - 7	Queens Pioneers	0.800
6 - 7	New Jersey Knockouts	0.800
8 - 9	Miami Sharks	0.633
8 - 9	Seattle Sluggers	0.633
10 - 12	Carolina Cobras	0.600
10 - 12	Tennessee Tempo	0.600
10 - 12	New York Knights	0.600

(Full disclosure: I am the team writer for the New Jersey Knockouts, although as I've said, the power rankings above have no subjective elements.)