What's new
Fantasy Football - Footballguys Forums

This is a sample guest message. Register a free account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members through your own private inbox!

FiveThirtyEight.com - Nate Silver's site (1 Viewer)

I like the premise of "data journalism" and taking a sabermetric look at all things in life.

Reminds me a bit of Isaac Asimov's "Foundation" series with Hari Seldon putting mathematical equations to history and sociology to determine the fall of the empire and planting the seeds to eventually rebuild it per his equations. Will we be able to do this on a smaller scale someday? :nerd:

 
This looks atrociously bad so far. "Teams that have bad passing attacks tend to draft quarterbacks early in the draft." So glad I clicked that link.

The internet already has one Slate. Not sure we need another.

 
His NCAA picks/analysis are straight chalk across the board. I get that's his shtick but uhhhhh....it's not interesting?

 
IvanKaramazov said:
This looks atrociously bad so far. "Teams that have bad passing attacks tend to draft quarterbacks early in the draft." So glad I clicked that link.

The internet already has one Slate. Not sure we need another.
Click to expand...
If that's all you got out of it you missed the point about as much as possible.

 
NCCommish said:
IvanKaramazov said:
This looks atrociously bad so far. "Teams that have bad passing attacks tend to draft quarterbacks early in the draft." So glad I clicked that link.

The internet already has one Slate. Not sure we need another.
Click to expand...
If that's all you got out of it you missed the point about as much as possible.
Click to expand...
The mock drafts available at ESPN.com — Mel Kiper’s and Todd McShay’s, for example — suggest that the Cleveland Browns will probably select a quarterback with the fourth pick in this year’s NFL draft. In Kiper’s view, the Oakland Raiders will pick a quarterback early, too.

If any of this sounds familiar, it’s because the Browns took a quarterback in the first round two years ago, and spent a third-rounder on one in 2010, and used another first-round pick on a passer in 2007. The Raiders took quarterback Tyler Wilson in the fourth round a year ago. They also, infamously, spent the first overall pick on JaMarcus Russell in 2007.

If a team’s in the market for a quarterback early in the 2014 draft, the odds are good that it has picked one highly in the recent past (the Houston Texans being the exception).

If we assume the Browns and Raiders both take passers in this year’s top five, then no other franchises will have devoted more of their draft resources (according to FootballPerspective.com’s draft value chart) to the quarterback position over the past decade. But for all the investment in passers, those two franchises were also among the least efficient passing teams (as measured by adjusted net yards per attempt) over the past 10 seasons.

Team Draft Pts Spent on QBs Adjusted Net YPA OAK 47.4 4.85 CLE 46.5 4.54 NYJ 35.7 4.80 MIN 34.7 5.42 MIA 34.6 4.92 WAS 34.4 5.54 TEN 33.8 5.35 CAR 33.0 5.45 SFO 32.3 5.03 DET 32.1 5.25 DEN 29.6 6.59 STL 23.7 4.93 IND 23.6 6.87 ATL 18.6 5.78 PHI 18.3 6.26 TAM 18.3 5.30 GNB 18.0 6.77 ARI 16.0 5.19 BUF 15.7 4.88 BAL 13.5 5.26 JAX 13.3 5.29 CIN 10.6 5.68 SEA 10.2 5.61 NWE 8.8 7.18 DAL 7.0 6.39 KAN 6.7 5.38 PIT 6.1 6.30 SDG 6.0 6.96 CHI 6.0 4.82 NYG 5.7 5.76 HOU 2.0 5.88 NOR 1.9 6.80 It’s a bit of a chicken-or-egg dilemma. Teams who pass poorly will continually turn back to the draft to grab a potential franchise quarterback. But does investing early draft picks in quarterbacks make any difference to a team’s future passing performance? This needs a bit more research. I’m looking into it and will get back to you.

But as the Browns and Raiders have discovered, it’s easy to fall into a vicious cycle: Draft a highly touted quarterback, watch as he performs poorly, draft another quarterback, repeat ad infinitum.
Click to expand...
 
This paragraph strikes me as particularly egregious.

If we assume the Browns and Raiders both take passers in this year’s top five, then no other franchises will have devoted more of their draft resources (according to FootballPerspective.com’s draft value chart) to the quarterback position over the past decade. But for all the investment in passers, those two franchises were also among the least efficient passing teams (as measured by adjusted net yards per attempt) over the past 10 seasons.
Click to expand...
Why would you possibly count this year's picks (who have an adjusted net yards per attempt of N/A in the NFL) in the picks devoted to quarterbacks and then compare it to the historical adjusted net yards per attempt? The only reason I can think of is to have the Raiders and Browns end up 1 and 2 in both categories. That's not a rigorous statistical model. It's a blatantly misleading one.

 
NCCommish said:
IvanKaramazov said:
This looks atrociously bad so far. "Teams that have bad passing attacks tend to draft quarterbacks early in the draft." So glad I clicked that link.

The internet already has one Slate. Not sure we need another.
Click to expand...
If that's all you got out of it you missed the point about as much as possible.
Click to expand...
It's a pretty poor article that fails to make any sort of point at all. Obviously teams with bad QB's are going to continue to pick QB's and so they will spend more draft picks on QB's than teams with good QB's.

 
Ramsay Hunt Experience said:
This paragraph strikes me as particularly egregious.

If we assume the Browns and Raiders both take passers in this year’s top five, then no other franchises will have devoted more of their draft resources (according to FootballPerspective.com’s draft value chart) to the quarterback position over the past decade. But for all the investment in passers, those two franchises were also among the least efficient passing teams (as measured by adjusted net yards per attempt) over the past 10 seasons.
Click to expand...
Why would you possibly count this year's picks (who have an adjusted net yards per attempt of N/A in the NFL) in the picks devoted to quarterbacks and then compare it to the historical adjusted net yards per attempt? The only reason I can think of is to have the Raiders and Browns end up 1 and 2 in both categories. That's not a rigorous statistical model. It's a blatantly misleading one.
Click to expand...
Wow. Yeah, I didn't even notice that.

 
Huge fan of 538.com both before the NYT and after, but the new page was really disappointing -- both the presentation and the subject matter.

 
shader said:
NCCommish said:
IvanKaramazov said:
This looks atrociously bad so far. "Teams that have bad passing attacks tend to draft quarterbacks early in the draft." So glad I clicked that link.

The internet already has one Slate. Not sure we need another.
Click to expand...
If that's all you got out of it you missed the point about as much as possible.
Click to expand...
It's a pretty poor article that fails to make any sort of point at all. Obviously teams with bad QB's are going to continue to pick QB's and so they will spend more draft picks on QB's than teams with good QB's.
Click to expand...
Yeah, an actually useful analysis would be looking at whether drafting a QB early in the first round on average has a better return on passing stats than drafting one later and historically how much you improve a bad passing attack via doing so versus later picks.

 
I don't get the Nate Silver love. People were going nuts about how he picked "all 50 states" right last election, but come on. I think everyone here could have probably picked 42 right, before we even got to the hard ones. And even those 8 were pretty easy to call with the exception of about 2 of them. So that means he had to pick 2 states right. If we all did that, I bet a lot of us would have picked all 50 states right, too. It really wasn't that amazing.

Unless I'm missing something. Which I usually am. And then I end up looking like an idiot. A lot.

 
TheIronSheik said:
I don't get the Nate Silver love. People were going nuts about how he picked "all 50 states" right last election, but come on. I think everyone here could have probably picked 42 right, before we even got to the hard ones. And even those 8 were pretty easy to call with the exception of about 2 of them. So that means he had to pick 2 states right. If we all did that, I bet a lot of us would have picked all 50 states right, too. It really wasn't that amazing.

Unless I'm missing something. Which I usually am. And then I end up looking like an idiot. A lot.
Click to expand...
Nate Sliver benefitted a lot from statistical illiterates in the GOP and at Politico. In a normal world, pooling a handful of polls to get a lower standard error than any one individual poll would be Statistics 201 stuff. Fortunately for Silver, a large proportion of people in the political realm never made it to Stat 201.

 
TheIronSheik said:
I don't get the Nate Silver love. People were going nuts about how he picked "all 50 states" right last election, but come on. I think everyone here could have probably picked 42 right, before we even got to the hard ones. And even those 8 were pretty easy to call with the exception of about 2 of them. So that means he had to pick 2 states right. If we all did that, I bet a lot of us would have picked all 50 states right, too. It really wasn't that amazing.

Unless I'm missing something. Which I usually am. And then I end up looking like an idiot. A lot.
Click to expand...
But he used math to statistically prove his predictions. He is the modern genius.

 
TheIronSheik said:
I don't get the Nate Silver love. People were going nuts about how he picked "all 50 states" right last election, but come on. I think everyone here could have probably picked 42 right, before we even got to the hard ones. And even those 8 were pretty easy to call with the exception of about 2 of them. So that means he had to pick 2 states right. If we all did that, I bet a lot of us would have picked all 50 states right, too. It really wasn't that amazing.

Unless I'm missing something. Which I usually am. And then I end up looking like an idiot. A lot.
Click to expand...
Nate himself says this in his opening manifesto on the new site.

He didn't do appreciably better than other statistical models (he did a little better). He did a lot better than mainstream journalists.

The point isn't whether Nate Silver is the analyst to go to, it's whether data-driven analysis is better than anecdotal/conventional wisdom driven analysis (or in the case of 2012, wishful thinking driven analysis).

 
IvanKaramazov said:
TheIronSheik said:
I don't get the Nate Silver love. People were going nuts about how he picked "all 50 states" right last election, but come on. I think everyone here could have probably picked 42 right, before we even got to the hard ones. And even those 8 were pretty easy to call with the exception of about 2 of them. So that means he had to pick 2 states right. If we all did that, I bet a lot of us would have picked all 50 states right, too. It really wasn't that amazing.

Unless I'm missing something. Which I usually am. And then I end up looking like an idiot. A lot.
Click to expand...
Nate Sliver benefitted a lot from statistical illiterates in the GOP and at Politico. In a normal world, pooling a handful of polls to get a lower standard error than any one individual poll would be Statistics 201 stuff. Fortunately for Silver, a large proportion of people in the political realm never made it to Stat 201.
Click to expand...
Doesn't the bold drive you nuts as a conservative? Especially since much of their policy is driven/support by their statistical illiteracy?

 
tommyGunZ said:
IvanKaramazov said:
TheIronSheik said:
I don't get the Nate Silver love. People were going nuts about how he picked "all 50 states" right last election, but come on. I think everyone here could have probably picked 42 right, before we even got to the hard ones. And even those 8 were pretty easy to call with the exception of about 2 of them. So that means he had to pick 2 states right. If we all did that, I bet a lot of us would have picked all 50 states right, too. It really wasn't that amazing.

Unless I'm missing something. Which I usually am. And then I end up looking like an idiot. A lot.
Click to expand...
Nate Sliver benefitted a lot from statistical illiterates in the GOP and at Politico. In a normal world, pooling a handful of polls to get a lower standard error than any one individual poll would be Statistics 201 stuff. Fortunately for Silver, a large proportion of people in the political realm never made it to Stat 201.
Click to expand...
Doesn't the bold drive you nuts as a conservative? Especially since much of their policy is driven/support by their statistical illiteracy?
Click to expand...
The GOP has people to listen to. The question is becoming whether or not they do so.

 
Ramsay Hunt Experience said:
TheIronSheik said:
I don't get the Nate Silver love. People were going nuts about how he picked "all 50 states" right last election, but come on. I think everyone here could have probably picked 42 right, before we even got to the hard ones. And even those 8 were pretty easy to call with the exception of about 2 of them. So that means he had to pick 2 states right. If we all did that, I bet a lot of us would have picked all 50 states right, too. It really wasn't that amazing.

Unless I'm missing something. Which I usually am. And then I end up looking like an idiot. A lot.
Click to expand...
Nate himself says this in his opening manifesto on the new site.

He didn't do appreciably better than other statistical models (he did a little better). He did a lot better than mainstream journalists.

The point isn't whether Nate Silver is the analyst to go to, it's whether data-driven analysis is better than anecdotal/conventional wisdom driven analysis (or in the case of 2012, wishful thinking driven analysis).
Click to expand...
He does. He even cites to people who got 48 or 49 states using stats.

 
Ramsay Hunt Experience said:
TheIronSheik said:
I don't get the Nate Silver love. People were going nuts about how he picked "all 50 states" right last election, but come on. I think everyone here could have probably picked 42 right, before we even got to the hard ones. And even those 8 were pretty easy to call with the exception of about 2 of them. So that means he had to pick 2 states right. If we all did that, I bet a lot of us would have picked all 50 states right, too. It really wasn't that amazing.

Unless I'm missing something. Which I usually am. And then I end up looking like an idiot. A lot.
Click to expand...
Nate himself says this in his opening manifesto on the new site.

He didn't do appreciably better than other statistical models (he did a little better). He did a lot better than mainstream journalists.

The point isn't whether Nate Silver is the analyst to go to, it's whether data-driven analysis is better than anecdotal/conventional wisdom driven analysis (or in the case of 2012, wishful thinking driven analysis).
Click to expand...
It $eem$ like he did appreciably better than other $tati$tical model$.

 
tommyGunZ said:
IvanKaramazov said:
TheIronSheik said:
I don't get the Nate Silver love. People were going nuts about how he picked "all 50 states" right last election, but come on. I think everyone here could have probably picked 42 right, before we even got to the hard ones. And even those 8 were pretty easy to call with the exception of about 2 of them. So that means he had to pick 2 states right. If we all did that, I bet a lot of us would have picked all 50 states right, too. It really wasn't that amazing.

Unless I'm missing something. Which I usually am. And then I end up looking like an idiot. A lot.
Click to expand...
Nate Sliver benefitted a lot from statistical illiterates in the GOP and at Politico. In a normal world, pooling a handful of polls to get a lower standard error than any one individual poll would be Statistics 201 stuff. Fortunately for Silver, a large proportion of people in the political realm never made it to Stat 201.
Click to expand...
Doesn't the bold drive you nuts as a conservative? Especially since much of their policy is driven/support by their statistical illiteracy?
Click to expand...
Should someone who aligns himself with the party that gave us Obamacare be taking jabs at others about statistics?

You might want to reign that in a bit.

 
The problem with the new 538 model is that it relies on quality contributions from people outside of Silver itself less it dilutes the product too much. So far, this quality has been pretty poor. It is the same issue that happened with Ezra Klein as he expanded his "brand" over the past few years. This broader focus will further hinder it. Particularly with all the linkbait going on.

 
ShaHBucks said:
shader said:
http://fivethirtyeight.com/datalab/how-to-eat-at-mcdonalds-when-youre-monumentally-broke/

I have no idea why this article measures dollar per gram. Dollar per calorie would be far more interesting. Either way, article that shouldn't have interested me, but did.
Click to expand...
That reminded me of the saying "pennywise and pound foolish." Some people like to pay what they weigh, I guess.He did cal/$ at the end of the article. It's basically the whole dollar menu.
Click to expand...
high-end analytics here, i guess.

certainly not the sort of thing i was expecting from a Nate Silver project.

 
The online reviews right now are overwhelmingly negative. That McDonald's article seems like a Stats I project.

 
Premier said:
The online reviews right now are overwhelmingly negative. That McDonald's article seems like a Stats I project.
Click to expand...
The political and economics articles seem to have more meat to them, but overall the collection of articles seems more like a quickly thrown together statistical analyses to fill out the site. Hopefully things grow from here.

 
shader said:
Ramsay Hunt Experience said:
This paragraph strikes me as particularly egregious.

If we assume the Browns and Raiders both take passers in this year’s top five, then no other franchises will have devoted more of their draft resources (according to FootballPerspective.com’s draft value chart) to the quarterback position over the past decade. But for all the investment in passers, those two franchises were also among the least efficient passing teams (as measured by adjusted net yards per attempt) over the past 10 seasons.
Click to expand...
Why would you possibly count this year's picks (who have an adjusted net yards per attempt of N/A in the NFL) in the picks devoted to quarterbacks and then compare it to the historical adjusted net yards per attempt? The only reason I can think of is to have the Raiders and Browns end up 1 and 2 in both categories. That's not a rigorous statistical model. It's a blatantly misleading one.
Click to expand...
Wow. Yeah, I didn't even notice that.
Click to expand...
How is he mixing the 2 statements? I think he's saying if they pick QBs this year, they will have used the most draft resources on QB from 2005-2014 even though they've had the worst QB stats from 2004-2013.

But I agree that article doesn't have much of a point. They are repeatedly trying (and failing) to find a good QB. no kidding

 
I love the idea of the site but it looks like right now their process is:

1. Hold a brainstorming meeting to think of potentially insightful articles based on stats

2. Assign accepted ideas to the writers, have them go off for a week and come up with something

3. Post whatever they have, even if the conclusion is obvious

I think for a site like this you need a bigger staff and a filtering process before publishing. They're going to need to reject most of the ideas researched for days that lead to a dead end or obvious conclusion. You need more than 20 people to do that.

 
The "articles" you guys are looking at are not actually articles. The stuff under "Datalab" comprises unedited, half-baked thoughts.

For the articles, go to the bottom under politics, economics, sports, etc., not datalab.

The layout is bad.

 
17seconds said:
I love the idea of the site but it looks like right now their process is:

1. Hold a brainstorming meeting to think of potentially insightful articles based on stats

2. Assign accepted ideas to the writers, have them go off for a week and come up with something

3. Post whatever they have, even if the conclusion is obvious

I think for a site like this you need a bigger staff and a filtering process before publishing. They're going to need to reject most of the ideas researched for days that lead to a dead end or obvious conclusion. You need more than 20 people to do that.
Click to expand...
If it is taking a week to come up with these POS articles, they have bigger problems than needing a filtering process.

Quality > quantity which is what made 538 such an interesting read in the first place. I was hoping leaving the NYT would get back to that, should have known Disney would have his site full of superficial garbage.

 
Maurile Tremblay said:
The "articles" you guys are looking at are not actually articles. The stuff under "Datalab" comprises unedited, half-baked thoughts.

For the articles, go to the bottom under politics, economics, sports, etc., not datalab.

The layout is bad.
Click to expand...
It's right on the front page. Not my fault they put garbage above the fold (so to speak).

 
http://fivethirtyeight.com/features/fivethirtyeight-senate-forecast/

FiveThirtyEight Senate Forecast: GOP Is Slight Favorite in Race for Senate Control

By Nate Silver

When FiveThirtyEight last issued a U.S. Senate forecast way back in July we concluded the race for Senate control was a toss-up. That was a little ahead of the conventional wisdom at the time, which characterized the Democrats as vulnerable but more likely than not to retain the chamber.

Our new forecast goes a half-step further: We think the Republicans are now slight favorites to win at least six seats and capture the chamber. The Democrats position has deteriorated somewhat since last summer, with President Obamas approval ratings down to 42 or 43 percent from an average of about 45 percent before. Furthermore, as compared with 2010 or 2012, the GOP has done a better job of recruiting credible candidates, withsome exceptions...

 
The long knives are out for Silver right now. He's getting destroyed by the reviews.

Some of it is predictable old-media backlash against a young guy who turned his back on the NYT after making big demands, and some of it is from people who probably never liked him in the first place (FOX News types), but the criticism is mostly pretty fair so far IMO.

He's off to a bad start.

 
Last edited by a moderator:
You must log in or register to reply here.

Similar threads

Max Power
Dating Shows
Replies
36
Views
1K
Runkle
Runkle
matttyl
Any electricians here? Nearly had electrical fire yesterday
Replies
27
Views
558
The Z Machine
The Z Machine
Joe Bryant
Small Scale Farming - Anyone Into It?
2
Replies
60
Views
1K
Chaos34
Chaos34

Users who are viewing this thread

Total: 1 (members: 0, guests: 1)
Back
Top