jimdahl

See this thread -- AZSioux beat you to it

Gophers were robbed, they had the goal.

Not for me As penance for humor-impairment this afternoon, I'll share one more interesting effort. There's been a lot of speculation about what UND needs to do to make the tournament. Here's what UND's PWR could look like at the end of the regular season, based on how many remaining games they win. Obviously these ranges are much broader and much more subject to change than a single weekend forecast: A PWR of 13 is possible with winning 60%, a tossup with 70%, becomes likely with 80%. I'm done for the week. Someone can now jump in and post "just win!"

Gee, I thought you actually made a really important point. I guess it was accidental

Your implied point that the games haven't been played yet and outcomes can be whatever the team makes them is an important one, and actually a key element of the forecasts. That's why the output is multiple distributions of possible PWR outcomes instead of a single PWR. They're not predictions of game outcomes, they're "what ifs" that forecast what various potential outcomes would do to PWR. A simulation that predicted game outcomes by some existing ranking (e.g. KRACH) would produce a set of PWR rankings consistent with current KRACH expectations. In early January KRACH predicted that UND would lose 3 of the 4 vs. Minnesota and Denver. Those predictions were based on performance to date, and if true would have precluded UND from making the tournament (which a lot of people at the time were guessing was impossible for the Sioux, based on intuition instead of actual calculations). Instead, I simulate across all the possible game outcomes to help answer the conditional questions (including those that deviate from current KRACH predictions) that people are continuously asking about PWR -- if UND defies expectations and sweeps the Gophers, what will that do to PWR? If UND wins out 70%, can it climb to the top 12 in PWR? If UND sweeps this weekend, what could its PWR be? etc...

The techniques are explained a little more deeply here and here, but in short: * I run 1,000,000 trials of the set of games in interest (either the remaining season or, as above, the next weekend) * Outcomes are either random (to simulate possible outcomes) or KRACH-based (to simulate likely outcomes) * Calculate the PWR for each trial, save, and aggregate Simulating every combination of outcomes would be tough. With 49 games this weekend, that would be 2^49, or about 562,949,953,421,312 possibilities. I've done some informal sensitivity analysis, and 1,000,000 trials does a pretty good job of producing consistent results. For the chart embedded above, I use KRACH-based likely outcomes (hence no ties) with fixed UND outcomes because I think that's the best determinant of what UND's PWR is likely to be after the weekend. I can also do the same thing for the entire season, as shown in the original blog post (the first "here" above). The cool thing is, a single run produced similar data for every team, so once I make it an automated process that runs itself and produces a web page, I should be able to make a page for every team that includes: outlook for the team for the weekend and key non-team games this weekend (see my blog post for an example).

Having last week's chart in the thread keeps me honest, but I declare victory on that, with 3 points we landed between the thick parts of the forecast for a sweep and a split. (Keep in mind that when I give a "90% chance" range, we are by definition going to land outside of that range 1 out of 10 times). From Forecasting the PWR… a look ahead at SCSU: One point of clarification -- the numbers on the bottom are the PWR ranking, not the number of comparisons won. Being further left (closer to 1) is better. That's why the far left curve is UND sweeping and the far right curve UND getting swept.

Given how close Duluth's performance has been to be UND's, I might have expected them to be close in PWR. Duluth has a slightly higher Win% than UND (.6154 vs. .5893), and a better RPI (.5410 vs .5365). Their competition was weaker (.5075 opp win% vs. .5309 opp win%) but their opp-opp win% are much closer (.5196 vs. .5143). It's that last fact that really goes against your intuition about who has played a tough schedule. In 2007 the RPI really deemphasized actual opponents' winning percentage in favor of opp-opp win%, which is much flatter. I think it's a horrible change, an opinion which I tried to justify in this Jan. 07 post. In trying to eliminate games in which winning would hurt your RPI, they had to deemphasize strength of schedule to the point that the negative effect on SOS of playing, say American International, is smaller than the gain in winning percentage from defeating them. SOS was neutered.

box score

Yep, I just used the inflammatory headline from the Forum to cause trouble. When you read the text, it's actually reasonable. However, the exaggerated version plays so well into some Chapman stereotypes that it's an irresistible opportunity for humor.

This article, Caterpillar to slash 20,000 jobs as profit falls, got me thinking about Deere and its Phoenix Intl unit in Fargo. While Deere recently reported great results, that was for FY08, which ended Oct. 31. The Caterpillar article noted that their business really hit the wall in Dec. I wonder if Deere will see any of the same weakness? Edit -- forgot, Bobcat too, right?

New NDSU security post guards Chapman, officials There are so many things that could be said here, but I think I'll just leave it with the link to the story.

In the 2001 and 2002 seasons, Bochenski scored a bomined 52 goals. (Men's hockey, of course).

Jan. 24 1253 -- Ryan Chappell's rushing yds in 2006

Yeah, Whelan is THE original PWR guru. That same site, circa 2000, was the first to break out the individual comparisons which allowed us all to begin hypothesizing about which comparisons could flip. No doubt it inspired me to build the PWR and RPI details features here, and him having nailed the You-Are-The-Committee (now at CollegeHockeyNews) and build-a-PWR (linked above) tools are why I've never bothered trying to duplicate either of those. I should work some links to them into the PWR screens somehow... That said, we disagree on something, in that USCHO and I have BC landing at #12 after defeating Maine. Perhaps you excluded today's other games? That's always been the problem with the build your own scenarios, one result can invalidate them. That's a big part of why I'm trying to predict distributions of outcomes instead of specific outcomes now.

Yeah, see the Rocky Mountain News article:

I'm missing it -- what's different? USCHO SiouxSports I still disagree with the pessimism in this thread. Three weeks ago I tried to debunk the "UND can't make the tournament" discussions, noting that winning about 75% of their remaining games would land them in the top 15 of the PWR. Here they are 6 games later having gone 4-0-2, already climbed to #15 in the PWR with 10 games remaining in the regular season, and the pessimism only seems to have gotten worse! At a quick glance, about half of UND's comparisons look flippable with a ~70% win percentage (which would land UND around a .542 RPI). Keep in mind that 10 games is also just the end of the regular season. The conference tournament can provide up to 6 more games (though realistically, a max 4-0-0 contribution for our purposes). In addition to an opportunity to move the RPI somewhere in the neighborhood of .01-.02, it can also present golden opportunities to secure additional h2h points against the likes of Denver, Minnesota, and UMD (each of whom has a tenuous 0-1 point lead over UND). I'll run some real numbers tomorrow, but at first glance my guess would have been that the Sioux could make the tournament at large by winning something in the neighborhood of 60-70% of its remaining games. I even wonder if the Sioux could land in the neighborhood of #8 in the PWR by the end of the regular season if they keep up their 2009 performance to date (a goal that would seem entirely unrealistic if last year's team hadn't done it).

I had the same thought -- these were the letters he got IN FAVOR of UND's move? JBB!?! I clearly have my own (self-serving?) bias, but reading that left me very pleased that fans can now get and share info without it being filtered and vetted. If the Forum/Herald were to be believed, all Sioux fans strongly oppose the move to D-I and the Fighting Sioux nickname.

From the Saturday night broadcast (somewhat paraphrased):

I'll try this with the short one, since Bob already linked the full version. Video embedding (instructions here): ">" type="application/x-shockwave-flash" width="425" height="350"> Of course, I still strongly encourage people to use the videos page to bring all interesting videos into one place, but it's also nice to be able to embed one into a relevant spot in a conversation.

Sure hope things go well in the "3rd frame"

Same for me, it doesn't come on until 9:00ET, but then only runs for two hours. Makes me think it's not even delayed, but just joined in progress. Oh, and since we're comparing beer stocks -- my fridge seems to contain 12 Winter Lagers, about 8 Sam Adams Boston Lagers, and 1 Red Stripe

40-50 cuts in Microsoft Fargo yesterday Though with only ~1000 of the planned ~5000 cuts in so far, who knows what the future holds.

Yeah, busy week. Now that I think I'm getting close to some readable, useful analysis of the forecasts, the next step will be to automate it so its not dependent on me having the time to run the simulation and type up the results.

We're only a month or so away from the Bracketology articles starting to get interesting and after a few weeks of playing around, I think I'm getting close to some useful ways to present PWR forecasts. So, without further ado: Predicting the PWR… The impact of the Denver series For those who can't be bothered to follow a link -- surprise, surprise, this is the biggest series of 2009 so far for PWR. A sweep will land UND an average of 8.5 spots higher in the PWR than getting swept. Other series this weekend that could move UND more than 1 spot: Northern Michigan over Alaska, Bemidji State over Niagara, SCSU over CC