clock menu more-arrow no yes mobile

Filed under:

Reviewing the Previews: Timo Seppa & VUKOTA Projections at Puck Prospectus

Now, this preview really is something you're not going to see much elsewhere.  When we think of stats for a player, we'll consider points, points per game, goals scored, etc. Because we can't see everyone play all the time, stats are useful to determine whether a player has done well or not.    But over the years there has been a growing undercurrent of statistical analysis available for hockey.   One of my favorites, and a site I use often for a deeper analysis is Behind the Net, which breaks stats down to when players are on/off the site as well as providing per-60 minutes numbers.

There's a larger site that reflects the growing use of statistical analysis: Puck Prospectus. The hockey cousins to the statistically-dominant Baseball Prospectus, they've really broke out with values such as GVT, GVS, and especially VUKOTA.  They not only break down how a player is currently doing, but VUKOTA actually looks to predict a player's performance by looking multiple variables such as past performance and comparable players.  As a result, they have been using this and other statistical methods to rank all 30 teams going into the coming season.

Now, this reference page really helps explain what the various abbreviations at Puck Prospectus reflects. However, the group of writers provides the results and explains them well in their VUKOTA-assisted team previews. Good if you really want to just get to the heart of the matter.  Timo Seppa was the author for the New Jersey Devils preview, and read on for a quick breakdown.

Now, Puck Prospectus' VUKOTA projections were referenced on ESPN's team previews but you get the whole thing right here and for no cost.  Right away from the headline, you'll notice that their system found the Devils to be the 11th best team in the league that year.  At the beginning of the preview, you'll see a break down of team statistics as well as a probability of the Devils' chances in the post season.  Much unlike Burnsides' preview, it doesn't come with a clause of "we keep predicting this" and it's definitely not cliche.  Because of their unique method, it makes one want to read on to find out what else is in there.  Especially if you're statistically inclined or you just want to read something completely different from most other team previews.

Initially, VUKOTA (and a Bayesian rating, these guys aren't messing around) predicts the team will take a step back next season in all facets of the game.  Yet, the overall ranking prevents one from thinking it's a gigantic failure.  Our guide through the predicted numbers, Timo Seppa, actually doesn't acknowledge these numbers - and it's for the best in my opinion. They really speak for themselves.  Instead, Seppa notes that the team's success is contingent on management and highlights the importance of Jacques Lemaire.   It's not a complete picture, but it can be forgiven as it is a really quick summary and really doesn't have much of a bearing on the most of the preview.

Puck Prospectus presents the predictions first by forward, by defensemen, and then by goalie before an overall conclusion.  I wish the writers in each of these at least linked to that reference page to remind/explain to the reader what GVT - Goals Versus Threshold - represents.   Basically, it represents what value in goals a player represents over a hypothetical replacement player. This is a big factor in who is good and who isn't, incidentally.

For forwards, Seppa is very praise-worthy and optimistic in light of the VUKOTA predictions.  And this really adds to the preview.  He - and the other Puck Prospectus writers - could have just listed the predictions and called it a day. But they are proper analysts at Puck Prospectus and Seppa admits that Zach Parise and Brian Rolston could very well exceed what is predicted of them; while proving how good Travis Zajac was last season and how important Jamie Langenbrunner and Patrik Elias will figure in 2009-10.  He even has a quick thought about other forwards like Dainius Zubrus (you'll probably love that Seppa thinks "he has a chance to put his money where his mouth is") and Brendan Shanahan.

Seppa only briefly goes over the major departures at forward, eventually concluding that "Lamoriello never seems to overpay and he always seems to have new faces waiting in the wings," though that may be a reference only to Bobby Holik. Still, it's an improvement over insinuations that losing Brian Gionta was a big deal.  Seppa goes another step further and has a whole paragraph on the forward prospects.  He notes that the Devils' best forward prospects are the three Swedes: Nicklas Bergfors, Mattias Tedenby and Jacob Josefson, the latter two who are still in Sweden and won't factor for another year or so.  Seppa thinks Bergfors is closer (more than just literally) and I agree with that point - it really is his time right now to make it.

There really isn't a lot to complain or argue with Seppa's additional editorial content about the forwards.  He covered all the bases from the big names right down to prospects.  The same goes for the defensemen.  He called the re-signing of Johnny Oduya as "the right choice" and even includes Mike Mottau as a solid defensemen (and the numbers actually support this to a point).  Seppa focuses a bit on Cory Murphy as a "Rafalski lite," noting that both did very well in the SM-Liiga.  Though Murphy has a real fight to make the team, it certainly builds some enthusiasm.  Seppa is quite right about Bryce Salvador's penalties being a net negative; and he may be right about Matthew Corrente not being able to make the team this year.    Seppa really has an insight to the Devils and truly did his research - which should be of no surprise since he is a statistician in his own right.

The preview ends with a small bit about two goaltenders and a "big picture" that acts as an overall view of the Devils. Seppa points out that the big question is whether Jacques Lemaire's coaching along with the Devils' prospects ability to replace those who have departed will be enough for the Devils to succeed.  I agree that it is a very valid question to ask, and I agree that the Devils have the history to suggest that the answer is yes.

Ultimately, this is an excellent, well-thought out, well-argued, and original preview. Seppa doesn't spend much of the article restating what the predicted numbers clearly state, even noting that they may be surpassed.  Again, my only real complaint is that you have to understand what the various GVTs and ESTRs represent to fully understand the VUKOTA predictions.  If you're new to all this, it may seem like Russian to you (ASIDE: Thanks to Jim at Blueshirt Banter, I found this Devils preview at that has Rich Chere and Tom Gulitti respond to questions about the team. Google translator doesn't do a good job with this article, so if you can parse it, by all means, take a crack at it in a FanPost.) A link to the Puck Prospectus reference page would have sufficed, I think, but that's a minor complaint.

Overall Outlook: Optimistic, Actual Prediction: 11th in the NHL (either 6th or 7th in the East at this time, where is their 16th ranked team?)

The Devils are predicted to be a little lower than I expect, but there's a method behind why - and that's what sets it apart from most other previews out there.  Seppa did such a good job writing the preview that you'll understand his thoughts and points regardless.  I highly recommend checking it out, it's the best Devils preview I've read so far. See if you agree or disagree with what Seppa has to say.  At a minimum, you'll learn something from it.