Nate Silver and Figure Skating | Golden Skate

Nate Silver and Figure Skating

Reginald

Match Penalty
Joined
Jan 28, 2011
Nate Silver of the New York Times has won national acclaim for his super accurate election predictions.

Are any of you very accurate at your figure skating predictions?

Does anybody think they're the best here?

Is there a Nate Silver of figure skating?
 

dorispulaski

Wicked Yankee Girl
Joined
Jul 26, 2003
Country
United-States
Whoever "Larry" is is very accurate :)

and definitely wins:

http://www.goldenskate.com/2007/09/interview-with-tessa-virtue-and-scott-moir/

Check Larry's predictions (in 2007!) for the 2010 Olympics in ice dance:

Larry (USA): OK, I’m going to go way out on a limb and predict the 2010 Olympic podium (in random order): Virtue and Moir, Davis and White, Domnina and Shabalin. Do you agree that these two teams are your biggest competition, or do you think that some of the old-timers will hang in there for three more years and challenge for medals?
 

gmyers

Record Breaker
Joined
Mar 6, 2010
Unfortunately there aren't enough people covering figure skating for there to be a Nate silver and many who Do cover don't really make predictions. Lot of people here are good but unlike elections there are many competitions with the same people in them so there are results to look at!
 

FSGMT

Record Breaker
Joined
Sep 10, 2012
OMG that was REALLY an incredible prediction! :eek: Especially considering that in 2007 V/M and especially D/W weren't really THAT high level (Meryl and Charlie's highest placement at Worlds was 7th)!! I won the GP predictions game this fall so I could consider myself quite good at predicitions... :biggrin:
 

Mrs. P

Uno, Dos, twizzle!
Record Breaker
Joined
Dec 27, 2009
To be fair, Nate Silver did incorrectly predict the Seattle Seahawks and the New England Patriots would face off in the Super Bowl. :p

Would have been great to see my home team there -- but love the whole Harbowl! :)
 

dorispulaski

Wicked Yankee Girl
Joined
Jul 26, 2003
Country
United-States
Yes, you are no slouch at predictions yourself, FSGMT!

Not only that, but Larry has all 3 in the right order!

In 2007 you would have had Tanith & Ben and Delobel and Schoenfelder ahead of D&W and DomShabs, although it was known DenStavs and DubLau were retiring.

Here's the Worlds 2007 top 10.

1 Albena DENKOVA / Maxim STAVISKI BUL 201.61 2 1 1
2 Marie-France DUBREUIL / Patrice LAUZON CAN 200.46 1 3 2
3 Tanith BELBIN / Benjamin AGOSTO USA 195.43 5 2 4
4 Isabelle DELOBEL / Olivier SCHOENFELDER FRA 195.19 4 5 3
5 Oksana DOMNINA / Maxim SHABALIN RUS 193.44 3 4 5
6 Tessa VIRTUE / Scott MOIR CAN 183.94 9 6 6
7 Meryl DAVIS / Charlie WHITE USA 179.14 10 8 7
8 Jana KHOKHLOVA / Sergei NOVITSKI RUS 178.29 6 7 8
9 Federica FAIELLA / Massimo SCALI ITA 170.75 7 9 11
10 Melissa GREGORY / Denis PETUKHOV USA 170.08 11 10 9
 

drivingmissdaisy

Record Breaker
Joined
Feb 17, 2010
I am the Natalie Silver of figure skating, and here's my bold prediction this year:

G: Kim
S: Sotnikova
B: Asada

You read it here first!
 

Buttercup

Record Breaker
Joined
Mar 25, 2008
That's an impressive prediction for sure ;)

Things I've called correctly: in the spring or summer of 2009, when skaters were rushing to return to competition ahead of the Olympics, I kind of suggested that Shen and Zhao will be the only comeback skaters to win an OGM. Unfortunately I cannot remember what thread this was in; possibly the massive Plushenko thread from a few years back, which means that I will never find it.

Also in 2009, when he was fresh off an 12th place finish at JW, I wrote that Yuzuru Hanyu was going to be a serious contender - but I missed the "when" about that, as he got better faster than I'd anticipated.
 

clairecloutier

Final Flight
Joined
Aug 27, 2003
I think Tim Goebel should become the Nate Silver of figure skating! Not only does he know skating, he's also a statistical analyst with a degree in mathematics. Sounds like a perfect fit. :)
 

Robeye

Final Flight
Joined
Feb 16, 2010
IMO, for a variety of reasons, figure skating as a sport is not ideally suited for Nate Silver-like predictions (and neither is the Super Bowl, btw). I'll only give further comments if anyone is actually actively interested. ;)
 

dorispulaski

Wicked Yankee Girl
Joined
Jul 26, 2003
Country
United-States
I can guarantee you that there is a Mathman who will surely read any post about predictions here ;)
 

Mrs. P

Uno, Dos, twizzle!
Record Breaker
Joined
Dec 27, 2009
IMO, for a variety of reasons, figure skating as a sport is not ideally suited for Nate Silver-like predictions (and neither is the Super Bowl, btw). I'll only give further comments if anyone is actually actively interested. ;)


I would tend to agree. I'd like to hear your thoughts!
 

Serious Business

Record Breaker
Joined
Jan 7, 2011
Some misconceptions about what Nate Silver does: His predictions are likelihoods, not certainties. He will tell people that having run a simulation of an election or a sporting event 1,000 times, 832 times the results were thus, and the other times the results were otherwise. In other words, he provides the odds.

Can this be applied to figure skating? Heck yes! Figure skating competitions generate a ton of data. There has to be a vigorous, holistic way to process that data to arrive at logical likelihoods of results. Now, here's the important part: it won't help you that much in predicting the total results of one competition. It might be able to tell you so and so is an odds-on favorite to win, but for a skater/team to have such a lopsided advantage, it would likely be obvious even without any number crunching. Where it would help is over the long term: if someone had such a prediction system in place and used it to bet on skating competitions against other people who don't have such a system, that person would make money in the long run. And perhaps more importantly, it would help skating federations strategize properly in deciding who to send to what competition to maximize results and placements. And again, that's something that would be yield positive results in the long run, if maximizing placements is the main goal of skater selection.

Some innovations/methodologies Nate Silver popularized with his election predictions that could be applied to skating:

Silver would weigh different polling firms' polls differently, according to their past track record. He'd also correct for any partisan lean in their past results. Prior to Silver, poll aggregators just averaged all the polls hodge podge, which wasn't particularly helpful. How can this be applied to skating? Some competitions are less reliable than others in predicting future results. We already know this, when it comes to the national competitions of a few country. But it might be useful to know just how biased they are, and how to extract useful info from them still. Some technical callers are harsher than others. So instead of simply saying something like this team had its dance pattern called a certain level X number of times out of Y, we can correct for the harshness of some tech callers and make a more informed prediction for how it'll be called in a future competition.

Nate Silver reminds people about the reversion to the mean constantly. When a candidate gets a spike in the polls after some news event, Silver is like, "calm down, everybody!" The polls will more likely than not go back to normal. In skating, skaters do have fluke competitions where they do unusually well. It's best not to point to that as an ongoing thing unless they actually keep it up for a few competitions. This also applies to things like season's best, or those dramatic list of the high scores in a season. A skater's highest score may not be a particularly useful data point for predicting how they'll do going forward.

Other things that would be interesting to look at, so we can actually quantify their effect:

Reputation PCS boost — We all know this happens. But by how much? How and what kind of exposure does a skater need to trigger it? How long does it last? Is it the same for all disciplines/nationalities?

Score inflation as a season progresses. For some reason, we see scores get higher and higher as the season goes on, more than can be explained by how much skaters have improved their programs. In some extreme cases, we even see poorer performances of a program at Worlds get a season's best. It'd be useful to quantify this effect. Does it apply to all skaters? Or just the well known ones? Can there be such a thing as scoring momentum for skaters independent of their progress? Can the size of the inflation be predicted based on the previous scores from the season if it is different from skater to skater?

Is any of this likely to happen in a public way, like it did with Nate Silver and his election predictions? Probably not. There just isn't that much public interest in skating. However, individual federations may do some of it on their own, without ever releasing the info to the public. Some of them may already do this in some form, to make better training and team selection choices.
 

gkelly

Record Breaker
Joined
Jul 26, 2003
Score inflation as a season progresses. For some reason, we see scores get higher and higher as the season goes on, more than can be explained by how much skaters have improved their programs. In some extreme cases, we even see poorer performances of a program at Worlds get a season's best. It'd be useful to quantify this effect. Does it apply to all skaters? Or just the well known ones? Can there be such a thing as scoring momentum for skaters independent of their progress?

If reputation scoring exists, and evidence is that it does at least on an unconscious level, I suspect that we would see the biggest effect in this direction from skaters who are having a breakout season. At first they would earn middle-of-the-road scores while they're still unknown, but as more judges become familiar with their work and as buzz begins to spread, and also as they earn the right to skate later in the skate order at seeded events, their scores would increase.
 

Robeye

Final Flight
Joined
Feb 16, 2010
I would tend to agree. I'd like to hear your thoughts!
Awright, you talked me into it :biggrin:.

(Caveat: this is in no way comprehensive or systematic, just some early impressions)

-Firstly, Serious Business gives an excellent explanation of what the output of Nate Silver's "prediction model" represents. It is a probability distribution of potential outcomes (more simply, what are the odds that any defined possibility will become reality?).

-Let's compare the subject of Nate Silver's most celebrated work, presidential elections, with things like figure skating championships (or the Super Bowl). What are some of the salient resemblances, and differences, and what are the ramifications for the analytical results?

The primary point of resemblance is pretty obvious: both elections and sporting events are contested activities, competitive pursuits. The primary purpose, PC niceties aside ("they are all winners..." etc. etc.), is to identify the champeen. One might therefore assume that the analytical expectations would therefore also be largely the same. But I would argue that such an assumption would be wrong.

Why is it that statistical analysts such as Nate Silver can often state that there is a better than 95% probability (and sometimes 99% and beyond), in other words, that it is a mortal lock, that so-and-so will win the election, and can say this before election day? If someone made a similar prediction for ISU Worlds or the Olympics (or the Super Bowl), and claimed it was an objective, probabilistic analysis, my advice would be: first, check to make sure your wallet is intact, and then run for the door as fast as you can. What accounts for this difference?

It's because the results of an election are to a large extent pre-ordained by the time we get to election day. There are two reasons for this. First, because it can be demonstrated that voters generally make up their minds before going to the ballot box, and the closer we get to the election date, this percentage starts to asymptotically approach 100%. Second, because of the trend toward increased early-voting, many actual votes have already been cast before the official Election Day. Thus, the votes, for all practical intents and purposes, have been locked in, and we know the score through robust polling and analysis.

In an election, this is why the projected winner's probability for success gradually increases in the days before the election; it basically reflects the fact (calculated from historical data) that the number of people who might change their minds has become too small to meaningfully alter the result, barring some bizarre game changer or Act of God.

This would be similar to doing a prediction for the outcome of, say, a best-of-seven-games contest such as the World Series, with the proviso that the prediction is iterated after the result of each game played. Like Nate Silver's election predictions, a team that won, for example, the first three games would see their probability rise to something very high, because 75% of their requirement for winning has been locked in.

In figure skating terms, it might be analogous to making predictions after the SP has already been skated, because roughly a third of the winning requirement has been locked in, and therefore the realistic final outcomes have started to narrow considerably.

In all three cases, the election, the World Series, the figure skating championship, if the forecast were made prior to any results being locked in, then they would all be much less predictively powerful, and possibly of a similar order of magnitude to each other (i.e. not that great).

One might argue that in skating, the "reputation" aspect of scoring is in some ways analogous to election votes being "locked in". But assuming for the sake of argument that reputation points exist, IMO such points are much more a "soft circle" phenomenon as compared to eve-of-election voter preferences, and they constitute a fairly small percentage of the total required to win. PCS may represent about a third of the points total, but not all of it can be assigned purely by reputation; the skater has to actually execute the program to a certain standard for a "reputation" bonus to come into play. I would think that such a rep bonus constitutes no more than 5-10% of the total score on the high side, again, assuming that it actually exists.

-While we may think that there are plenty of data points for making statistical predictions for podium results in figure skating, there actually aren't, in my view. Consider: in baseball, there are 160 or so games in the regular season, a wealth of data that can be mined. And this is a series of data that goes back, in recognizable form, for a century. Even so, it is difficult to make predictions about who will win the World Series based on regular season results (again, if one does iterative predictions after each World Series game, then it is a different story).

How many events are there in a skating season? A typical international schedule would be a couple of GP events, perhaps the GPF, possibly a Euro or 4CC, and then Worlds. Five events.

-A figure skating championship. like the Super Bowl, is a one-off rather than an extended series. It is not conceptually unreasonable to think that chance and fortune can possibly play a much greater role than for events in which a contestant has multiple bites at the apple.

-A skating competition is highly compressed relative to other sports. A few minutes on ice divided into two periods. A "clean" skate is the goal; a rough patch of ice surface, a momentary lapse of concentration or fatigue, even one such instance can, as a matter of course, put the top of the podium out of reach, in a way that does not often happen in a contest where each game lasts for several hours and you need to win four of them.

-This is why I admire figure skaters. The amount of mental fortitude required to accept and overcome such impossible demands in the pursuit of a championship engenders a certain amount of awe.
 

Buttercup

Record Breaker
Joined
Mar 25, 2008
I'm seriously disappointed that Nate Silver isn't a closet figure skating Svengali.
This would be similar to doing a prediction for the outcome of, say, a best-of-seven-games contest such as the World Series, with the proviso that the prediction is iterated after the result of each game played. Like Nate Silver's election predictions, a team that won, for example, the first three games would see their probability rise to something very high, because 75% of their requirement for winning has been locked in.

-While we may think that there are plenty of data points for making statistical predictions for podium results in figure skating, there actually aren't, in my view. Consider: in baseball, there are 160 or so games in the regular season, a wealth of data that can be mined. And this is a series of data that goes back, in recognizable form, for a century. Even so, it is difficult to make predictions about who will win the World Series based on regular season results (again, if one does iterative predictions after each World Series game, then it is a different story).
And of course, Silver started out writing baseball analysis - sabermetrics and such - for Baseball Prospectus. Baseball, because of the large sample sizes and the nature of the game, pretty much invites statistical analysis*. Figure skating, not as much.

* though more at the individual than the team level.
 
Last edited:
Top