Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - garygreen

Pages: [1] 2 3 ... 77  Next >
1
Philosophy, Religion & Society / Re: Terrible Political Memes
« on: November 25, 2022, 05:39:14 PM »

2
Arts & Entertainment / Re: Official Sports Thread
« on: November 16, 2022, 11:19:19 PM »
yeah i think i might be able to whip something like that up. i could maybe make a gif that goes from week 1 to current that shows all the teams moving around wpa/epa space.

3
Arts & Entertainment / Re: Official Sports Thread
« on: November 15, 2022, 05:48:47 PM »
the state of the nfl going into week 11




4
Arts & Entertainment / Re: Official Sports Thread
« on: November 14, 2022, 12:12:18 AM »
lmao fuck the bills

5
Arts & Entertainment / Re: Official Sports Thread
« on: November 13, 2022, 12:21:37 AM »
here's my first shot at a prediction model. it's just a vanilla logistic regression trained on three features for every nfl game since 2002: pythagorean expectation for each team, and a binary variable for home/away. it performed well on held-out validation data, and the fact that it seems to track the vegas odds so closely is very encouraging.

the left panel shows you my model's prediction vs the chance to win implied by the moneyline odds. the right panel shows you the difference between the predicted odds and the implied odds — this is the per-unit expected value of a bet for each team. positive numbers are good. you can multiply a team's per-unit EV by the size of a bet to get the average return on that bet (e.g., 0.05 * $100 = $5, so you should expect an average return of $5 for that bet).


6
Arts & Entertainment / Re: Official Sports Thread
« on: November 06, 2022, 02:57:19 PM »
the state of the nfl going into today's games. go chefs.


7
Arts & Entertainment / Re: Official Sports Thread
« on: October 29, 2022, 04:43:26 PM »
i changed some of the aesthetics on these plots to make them easier to read. i also made a new plot that is pretty interesting imo. this is average team wpa vs average team epa across all non-special-teams plays this season.

one way to read this plot is to look at vertical slices of ~constant epa per play. for example, the raiders and the giants both have the same average epa per play, but the giants have positive wpa per play while the raiders have negative wpa per play. in other words, both the raiders and the giants are equally effective at playing football, but the giants are performing better in situations that have a greater impact on the outcome of the game. and that is nicely reflected in their relative win percentages.

i'll leave it up to the reader to decide if that disparity is the result of things like poor situational coaching, team chemistry, "clutch" factor, momentum, or simply random variation.


8
Arts & Entertainment / Re: Official Sports Thread
« on: October 26, 2022, 12:34:46 AM »
i made a goof. i incorrectly filtered out some plays on my last plot that skewed the results. here is a better set. for reference, wpa is contextual and considers every play, e.g. scoring a touchdown when you're tied increases your odds of winning more than scoring when you're down by 69. epa, however, does not care about context, e.g. scoring a touchdown always adds the same expected value. to reduce systematic error due to game situations, epa only considers plays where the score is close (within 2 scores) and time is not a factor (more than 2 minutes remaining).

in other words, epa is "how good are you at moving/stopping the handegg in typical situations," and wpa is "how good are you at improving your odds of winning handegg games."

why bother with two metrics? well, let's say patrick mahomes throws an interception to end a game against the bills (yikes!). wpa is sensitive to the fact that this interception ended the game, so it penalizes this pic more than one thrown in, say, the first quarter of the game. but is mahomes (or any quarterback) really more likely to throw a pic in the 1st quarter than the 4th? if you think so, then wpa is your stat. if, however, you think player performance does not drastically change with respect to situation, then you wanna look to epa.


9
Arts & Entertainment / Re: Official Sports Thread
« on: October 25, 2022, 05:44:16 PM »
here's the state of the nfl today. fyi, negative numbers are good for defense, so the upper right quadrant is the good quadrant, and the lower left is the bad one. the jets are probably frauds. philly's defense has to come back down to earth sometime, but their offense may be legit. the giants may be getting mildly lucky.



The primary peak sitting at ~ -0.75 EPA seemed a bit weird that it was so consistent

i was puzzled by that, too, but i guess when you think about all the incompletions and run/pass plays that only go for a few yards or less, probably most nfl plays have a slightly negative EPA. i think the other thing that's happening is that it's difficult to judge the relative heights of the primary peaks with respect to each team. i'd expect buffalo and kc to have smaller peaks there than, say denver and carolina.


10
Arts & Entertainment / Re: Official Sports Thread
« on: October 21, 2022, 04:47:59 PM »
thanks fam

i got the data from some shit called the nflverse, but i made the plot myself. lol most of my time has been spent tweaking the colors. it needs a caption for sure.

i'm most surprised that nfl passing across the board is dominated by low epa plays. the distributions in general are a lot more similar than i would have expected. like i think if you just showed me one of the distributions at random, i wouldn't have any idea if it was associated with a good offense or a bad one. i'm looking into the data more carefully to see if i'm making an error in the way i group them. lol that is entirely possible.

11
Arts & Entertainment / Re: Official Sports Thread
« on: October 20, 2022, 11:29:22 PM »
i found a treasure trove of high-quality nfl data. i made a plot. it's not especially informative, but it looks neat imo. go chefs.


12
i'm actually curious to know how cruise missiles help in a humanitarian crisis. was the crisis a lack of explosions?

13
call me crazy but i feel like if you declare war on a sovereign nation then you are the one who caused a war with a sovereign nation. especially if you follow it up by sending a few hundred thousand troops across their borders to start killing people and capturing cities and such.

"but they wouldn't let us dictate their foreign policy!!!!!!!!" is a justification, i guess, but it's kind of a bad one imo imho.

lol let me guess — you're a huge fan of neville chamberlain

14
yeah it's super weird that ukraine doesn't want to reward the nation that is illegally annexing them with a treaty to cede sovereign territory and let them dictate ukraine's foreign policy. so weird. i simply can't understand why z is being so unreasonable.

15
Arts & Entertainment / Re: Official Sports Thread
« on: October 08, 2022, 12:38:30 AM »
i am extremely here for watching the broncos collapse just four weeks into wilson's contract

16
Arts & Entertainment / Re: Just Watched
« on: September 28, 2022, 01:52:00 AM »
severance

i fucking loved every single thing about season 1. it was as spectacular as helly's pencil skirts.

17
Philosophy, Religion & Society / Re: The Queen
« on: September 12, 2022, 02:26:21 PM »
i realize i am of course doing major mental gymnastics here (probably because i just don't understand politics on your level), but do you think it's possible that these constitutional monarchies all function along similar lines because they share a common cultural and developmental heritage? like maybe the fact that they are all european is not a complete coincidence? i could be wrong, but it just seems like they're all pretty closely related to one another in space and time, so maybe we should be cautious about overgeneralizing from what is clearly not any kind of random sample.
I'm not sure what point you think you're making. Like, yes, political culture is a very important part of any political system that relies heavily on convention? Did that need to be pointed out?

perhaps i am misunderstanding your point in general — and of course please point it out if i am — but imo you seem to be making the claim that constitutional monarchies are very democratic because they always have impotent executives, and impotent executives are good. others have pointed out that this is not a necessary feature of a constitutional monarchy. we can easily imagine a constitutional monarchy with a strong executive.

your argument, then, was that all current constitutional monarchies have this feature, and we can/should generalize from this fact — after all, what are the odds that they would all share this feature by accident?

this brings us to my argument above: the fact that those governments all share a similar heritage/culture/time in history/etc means that we ought not generalize from their shared features. they aren't random and independent measurements. there are clear covariances.

18
Arts & Entertainment / Re: Official Sports Thread
« on: September 12, 2022, 12:13:25 AM »

19
Philosophy, Religion & Society / Re: The Queen
« on: September 09, 2022, 10:44:41 PM »
The UK is not the only democratic constitutional monarchy. There are six monarchies in the EU alone, and several more elsewhere in Europe. With the exception of the absolute monarchy of the Vatican, they all function along similar principles of limited power granted to the sovereign, whether through legislation or convention. It takes some mental gymnastics to suppose that they have all simply been fortunate for the past century or two.

i realize i am of course doing major mental gymnastics here (probably because i just don't understand politics on your level), but do you think it's possible that these constitutional monarchies all function along similar lines because they share a common cultural and developmental heritage? like maybe the fact that they are all european is not a complete coincidence? i could be wrong, but it just seems like they're all pretty closely related to one another in space and time, so maybe we should be cautious about overgeneralizing from what is clearly not any kind of random sample.

p.s. sorry in advance if my question is meaningless ;););) wink emoji

20
Philosophy, Religion & Society / Re: Trump
« on: September 05, 2022, 02:13:33 PM »
this is just a straight-up lie. an easily falsifiable lie. https://www.lawfareblog.com/can-trump-just-declare-nuclear-secrets-unclassified

Did you even bother to read through your link?

The author says that the matter is unclear. The author only points out that the law on nuclear secrets doesn't specifically give the president blanket power to declassify. The law does not specifically prohibit the existing declassification powers of the president. The law doesn't talk about it.

"there are shitloads of regulations regarding declassification that ostensibly apply to everybody, but the degree to which executive authority mitigates this has yet to be tested in court" is a very far cry from your false assertion that "the courts can't convict trump because there are no declassification standards that bind the president."

maybe you should have bothered to read the next paragraph:

Quote
Unlike National Defense Information, the procedures for identifying and declassifying Restricted Data are defined in statute, not in executive order. Whatever assumptions one might make about whether presidents need to follow their own executive orders (or the executive orders of previous presidents) or not thus get thrown out the window here as well. That doesn’t totally resolve the constitutional situation: Maybe you could try to argue that Congress doesn’t have the power to punish presidents for releasing Restricted Data, because that might interfere with their operations as commander-in-chief. Or, one could argue whether the Restricted Data clause is inherently unconstitutional, which as we’ve seen is not a new argument. Either way, it is a different issue at heart with Restricted Data than it is with National Defense Information.

Pages: [1] 2 3 ... 77  Next >