The site has developed an algorithm to measure the excitement of a sporting event. The main idea they have now is to let you know if a game you have DVR'ed is worth watching (without telling you the score), and when the best time is to start watching. The algorithm takes into account factors such as the game's pace, novelty, and how close the score is. The output is supposed to be the excitement an unbiased individual would get from the game.
They approached me back in November, after I wrote about the Wizards-Pistons game in Detroit for Bullets Forever. The game registered a perfect 100 score in the Thuuz system and it turns out that these guys can do a lot with the algorithm and statistics. I got them to do a breakdown of each player's contribution to the game's excitement. They have done their best work with basketball, but measure the excitement of other sports as well.
For the table below, the only heading needing explanation is 'Impact.' This is a measure of the number of plays a player makes that contribute directly to the game's outcome - I believe this is something close to points, assists, rebounds, etc. The reasoning behind this measure is that the excitement rating would not highly rate a player on a good team, because a team that increases their lead is boring, not exciting, so a measure is needed for these types of players. I hope that makes sense. Without further adieu, the numbers for the Wiz-Pistons game:
Washington vs. Detroit | 11/21/2010 | ||||
Player | Play Time | Impact | Excitement | Impact/Minute | Excitement/Minute |
Arenas | 42.80 | 43 | 35.32 | 1.00 | 0.83 |
Hinrich | 39.40 | 25 | -2.95 | 0.63 | -0.07 |
Thornton | 31.32 | 17 | -3.94 | 0.54 | -0.13 |
Blatche | 43.88 | 37 | -19.97 | 0.84 | -0.46 |
McGee | 42.27 | 39 | -42.98 | 0.92 | -1.02 |
Armstrong | 10.63 | 4 | -14.10 | 0.38 | -1.33 |
Young | 30.68 | 20 | 5.92 | 0.65 | 0.19 |
Booker | 9.15 | 4 | 0.99 | 0.44 | 0.11 |
Martin | 14.67 | 9 | -9.43 | 0.61 | -0.64 |
Not surprisingly, Gil was the Wizards most impactful and exciting player. The interesting thing about this game though, is that John Wall did not play. So what happens when you add JWow to the mix? Fortunately, the Wiz played another game just 2 days later that also was a 100 on the excitement scale, this time with the Future. Here are the numbers:
Philadelphia vs. Washington | 11/23/2010 | ||||
Player | Play Time | Impact | Excitement | Impact/Minute | Excitement/Minute |
Arenas | 41.98 | 21 | 114.67 | 0.50 | 2.73 |
Hinrich | 34.85 | 21 | 58.08 | 0.60 | 1.67 |
Thornton | 8.17 | 0 | 6.28 | 0.00 | 0.77 |
Blatche | 42.70 | 37 | 62.35 | 0.87 | 1.46 |
McGee | 45.85 | 61 | 22.88 | 1.33 | 0.50 |
Armstrong | 5.45 | 1 | -8.99 | 0.18 | -1.65 |
Wall | 35.93 | 33 | 135.84 | 0.92 | 3.78 |
Martin | 6.40 | 1 | 12.74 | 0.16 | 1.99 |
Young | 31.75 | 27 | 74.61 | 0.85 | 2.35 |
Booker | 11.87 | 9 | 7.30 | 0.76 | 0.62 |
Granted, it was a huuuge game for JWow, but look at his excitement rating! This is only 1 game, but the bottom line is that Wall was the most exciting player from any team in either game. This also gives a nice little comparison of how the team looks in terms of impact with Wall on and off the court. I obviously sat on this data way too long, and any comparison with Gil Arenas is now moot with him off the team. If you are wondering why the Wizards have more total excitement in the 76ers game, it is because the rest of the excitement is accounted for by the opposition, so the Wizards drove the excitement in game #2.
Thuuz is starting to do a lot more with all of their data, so I just wanted to give you a taste of what they do as I begin to post their data more regularly. I have discussed with them what types of stats and metrics would be most interesting and useful, so let me know any and all ideas. The next step is for Thuuz to get larger sample sizes for individual players so we can truly track who is exciting over the course of a season........that way we'll have proof when John Wall becomes the league's most exciting player.
I like the idea but am so confused as to where the rating for "Excitement" comes from. I kind of understand impact...kind of. But excitement? Please explain.
ReplyDeleteExcitement for a game can be quantified because we like:
ReplyDelete-High scoring games
-Close games
-Fast-paced games
-Novel games - as in something unique happens
-In basketball, dunks, three-pointers, blocked shots...
So Thuuz took those factors and a few others, and put them together to literally rank how exciting a game is on a scale of 1-100.
There can be different factors for different sports, but you can figure out most of the basic data going into the algorithm, such as the points I listed above. Make sense?
A 76ers vs Wiz game got a 100 on the excitement scale?? Seems a little flawed to me. Also can we please think of a better nickname for John Wall. JWow is the worst.
ReplyDeleteSorry for the hate, still think it's a cool post tho for sure.
We don't spend much time on graphs, but if you guys like the concept, you should check out Are You Watching This?!. We created the Instant Classic alert, and are the best at it.
ReplyDeleteThanks,
Mark
http://areyouwatchingthis.com
Yea, Jwow is terrible and a horrible reference to Jersey shore. I don't want him ever compared to that trash.
ReplyDeleteSorry for the belated response. That Wizards 76ers game was excellent, good or bad teams. That said, it points to one of the flaws in the system, that a well-played close game between the Heat and Lakers could achieve a similar rating to the Wizards-76ers game that is not well played.
ReplyDeleteAs for the 'Are you Watching This' site, if you can provide stats for me to use, I'd be much obliged and happy to give you pub. Let me know.
Nicknames. Jersey Shore is amazing, JWow is not, but it's short and all the other ones out there don't quite work. I like GameChanger, but it's too long, Great Wall of Chinatown is also too long - maybe just 'The Great Wall' will work. DWade already has Flash....