Discstats @Discstats

Bringing Analytics to Frisbee discstats.wordpress.com Chicago, IL Joined June 2016

Tweets

68
Followers

173
Following

273
Likes

25

Discstats @Discstats

8 years ago

First time I've ever seen duration and timeout data outside the @theAUDL, perhaps @USAUltimate should look towards adopting the same tracking methods! #WUCC2018

0 0 2 0 0

View Details

Discstats @Discstats

8 years ago

Hey, @WUCC2018 What app/system are you using to track your games? The game log pages (e.g. results.wucc2018.com/?view=gameplay…) are fantastic!

2 0 5 0 0

View Details

@_Jacoblong Unfortunately @USAUltimate doesn’t record who wins the flip or who starts on offense in their game logs -frustrating! But i did a basic numerical analysis a few years ago: bit.ly/2mk9kQX . Basically, breaks are so common that starting on O is a very small advantage

1 1 0 0 0

View Details

Discstats @Discstats

8 years ago

This is based on 922 games, with 103 of those games going to Universe. The historical data is interesting, but there is not enough data to draw a definitive conclusion about offensive efficiency on Universe Point.

0 0 1 0 0

View Details

Discstats @Discstats

8 years ago

Some follow-up from the Universe Point data from yesterday, which showed offenses historically scoring slightly more often on Universe Point than at other times. In games that go to UP, offenses are more efficient overall, but that rate goes down slightly on Universe Point.

1 0 4 0 0

View Details

Ultiworld @Ultiworld

8 years ago

Deep Look: @WUCC2018 Preview & Picks, Pro-Elite, Small Ball --> buff.ly/2LaJfyH

0 2 5 0 0

View Details

Discstats @Discstats

8 years ago

@BrewersStats Earlier I checked the Hold Rates filtering out blowouts and only keeping Close Games (score within 5), but there was no discernible difference. I think I will see one for UP games as well. I’ll also follow up next week with some 1st/2nd half splits, not sure what I’ll see there.

1 0 0 0 0

View Details

Discstats @Discstats

8 years ago

@BrewersStats This is a great point, someone made the same comment and prompted me to look into it. I should have an update tonight/tomorrow. Mainly, what is the Hold Rate specifically in UP games?

1 0 3 0 0

View Details

Discstats @Discstats

8 years ago

The average Offensive Hold Rate across all divisions from Nationals 2014-2018 is 65.4%. On smaller sample size of Universe Points, the Hold Rate is 69%. More research needed to confirm trend, but early results go against conventional wisdom that "anything can happen" on UP.

4 0 10 0 0

View Details

Discstats @Discstats

8 years ago

Error: The Total Games column has the number flipped for Men's Club (it should be 148) and Women's College (it should be 255). The rate stats are correct.

0 0 0 0 0

View Details

Discstats @Discstats

8 years ago

@Ultiworld @skydmagazine @theAUDL @SinTheFields

1 0 0 0 0

View Details

Discstats @Discstats

8 years ago

@bamasecs I totally agree those would be great, but are not widely available except for AUDL, which I don’t want to use. I’m trying to ultimately do as much as I can with what is available, because I don’t think that resource has been fully tapped yet.

0 0 2 0 0

View Details

Discstats @Discstats

8 years ago

@bamasecs Team that received the pull

1 0 0 0 0

View Details

Discstats @Discstats

8 years ago

Updated Offensive Hold Rate data, now including @USAUltimate Mixed Club Data! Close Games are within 5 points, inclusive.

0 0 6 0 0

View Details

Discstats @Discstats

8 years ago

@UltiDavid @USAUltimate Good question. I’ll have a whole blog post in a few days, I’ll make sure to have that included.

0 0 1 0 0

View Details

Discstats @Discstats

8 years ago

Updated Offensive Hold Rates with College and Club. Surprisingly (to me, at least), data reveals nearly identical values for each gender, regardless of division.

2 0 11 0 0

View Details

Discstats @Discstats

8 years ago

@ckerr4 Would be great to combine with Elo system (again I know Craig has done it) and be able to give live updates

0 0 0 0 0

View Details

Discstats @Discstats

8 years ago

@ckerr4 I had one based on a flat rate on hold % that i calculated a few years ago on a smaller sample size. Now with more data, I can see how that rate varies with division and with game situation. These will lead to small but not insignificant improvements.

0 0 0 0 0

View Details

Discstats @Discstats

8 years ago

@ckerr4 This is not the first time I’ve accidentally tread on @cpoeppelman ‘s previous work... I was literally setting up a Win Prob calculator as I saw this tweet!