Sports

NHL Playoff Predictor 08-09 Round 2

Alright, the first round of the playoffs is in the books, so lets review how the predictor did, shall we?


Boston_Bruins 0.814 0.634 0.458 0.308
Washington_Capitals 0.649 0.321 0.117 0.051
New_Jersey_Devils 0.603 0.339 0.143 0.070
Pittsburgh_Penguins 0.495 0.209 0.084 0.036
Philadelphia_Flyers 0.505 0.216 0.087 0.038
Carolina_Hurricanes 0.397 0.136 0.055 0.021
New_York_Rangers 0.351 0.073 0.027 0.008
Montreal_Canadiens 0.186 0.071 0.029 0.010


San_Jose_Sharks 0.706 0.474 0.276 0.145
Detroit_Red_Wings 0.720 0.435 0.227 0.110
Vancouver_Canucks 0.617 0.276 0.121 0.049
Chicago_Blackhawks 0.688 0.390 0.217 0.107
Calgary_Flames 0.312 0.118 0.044 0.014
St_Louis_Blues 0.383 0.124 0.042 0.012
Columbus_Blue_Jackets 0.280 0.083 0.031 0.009
Anaheim_Ducks 0.294 0.101 0.043 0.014

In summary, the predictor didn’t do a bad job, but it did get 3 series wrong giving it a 5/8 record starting off this playoffs. The biggest upset was Anaheim over San Jose — a series that San Jose apparently didn’t even show up for. Besides that, the other two series that the predictor got wrong were really close. Carolina managed to hit runner-runner to win their series by forcing a game 7, and then scoring two goals to tie and go ahead late in the 3rd period. Thanks for screwing up my playoff pools, Carolina! The other series, my predictor had Pittsburgh and Philly only a few percentage points apart, and so I’m not really that concerned that it picked the wrong side there.

Looking ahead, here’s the output for the next round:


Boston_Bruins 0.770 0.570 0.388
Washington_Capitals 0.506 0.166 0.072
Pittsburgh_Penguins 0.494 0.161 0.070
Carolina_Hurricanes 0.230 0.103 0.040


Detroit_Red_Wings 0.678 0.358 0.164
Vancouver_Canucks 0.405 0.190 0.073
Chicago_Blackhawks 0.595 0.337 0.159
Anaheim_Ducks 0.322 0.115 0.035

Boston has a huge edge going against Carolina, and Detroit has a similar edge against the ducks. But the other matchups are *much* closer. The program gives the slight edge to the Caps, but given how much trouble the Caps had with the Rangers, I wouldn’t totally be surprised to see the Penguins move on. The Canucks/Blackhawks series isn’t quite that close, and Chicago is certainly looking like a good option.

Herald

NHL Playoff Predictor 08-09

The NHL season is over for another year. Alas, my Oilers didn’t make it, but it still is a fun time to run the playoff predictor for another season. Here’s the numbers for the various teams. For an explanation, see here.


Boston_Bruins           0.814 0.634 0.458 0.308
Washington_Capitals     0.649 0.321 0.117 0.051
New_Jersey_Devils       0.603 0.339 0.143 0.070
Pittsburgh_Penguins     0.495 0.209 0.084 0.036
Philadelphia_Flyers     0.505 0.216 0.087 0.038
Carolina_Hurricanes     0.397 0.136 0.055 0.021
New_York_Rangers        0.351 0.073 0.027 0.008
Montreal_Canadiens      0.186 0.071 0.029 0.010


San_Jose_Sharks         0.706 0.474 0.276 0.145
Detroit_Red_Wings       0.720 0.435 0.227 0.110
Vancouver_Canucks       0.617 0.276 0.121 0.049
Chicago_Blackhawks      0.688 0.390 0.217 0.107
Calgary_Flames          0.312 0.118 0.044 0.014
St_Louis_Blues          0.383 0.124 0.042 0.012
Columbus_Blue_Jackets   0.280 0.083 0.031 0.009
Anaheim_Ducks           0.294 0.101 0.043 0.014

Playoff Predictor, Stanley Cup Finals, 2008

Well my predictor is sure doing better than I did in my playoff pools. Going into the conference finals, it was 11 for 12 for predictions, and it got both of the conference finals right as well!

Pittsburgh_Penguins 0.579 0.204
Philadelphia_Flyers 0.421 0.119

Detroit_Red_Wings   0.681 0.498
Dallas_Stars        0.319 0.178

So what’s left? The Stanley cup finals start tomorrow night and it should be a great series. So I gave my predictor one final spin to see what it says. It should come as no surprise that Detroit is heavily favoured by the program:

Pittsburgh_Penguins 0.295
Detroit_Red_Wings   0.705

While I’d be happy to say my program went 14/15, I’m still kinda rooting for the penguins. They are an exciting team to watch with a hell of a lot of talent at a very young age. Whoever wins though, the series should be one for the ages.

Heraldk

Playoff Predictor, Round 3 2008

Well the second round of the playoffs is over, and it’s time to give my simulator another spin. First, let’s look at the results from Round 2:

Montreal_Canadiens  0.616 0.348 0.147 Detroit_Red_Wings  0.772 0.530 0.380
Philadelphia_Flyers 0.384 0.174 0.056 Colorado_Avalanche 0.228 0.091 0.039

Pittsburgh_Penguins 0.568 0.289 0.114 San_Jose_Sharks    0.483 0.180 0.094
New_York_Rangers    0.432 0.189 0.064 Dallas_Stars       0.517 0.199 0.107

The simulator goes 3 for 4 this round with the only one not going its way being the Montreal-Philly series. Overall, the simulator’s predictions have been 11 for 12. Not too shabby huh?

Here are the predictions of my simulator for the conference finals:

Pittsburgh_Penguins     0.579 0.204
Philadelphia_Flyers     0.421 0.119

Detroit_Red_Wings       0.681 0.498
Dallas_Stars            0.319 0.178

So according to the program, Detroit has just shy of a 50% chance to win the whole thing, and their likely opponents is the Pittsburgh Penguins. Dallas and Pittsburgh are very close for 2nd place and Philly is not liked as much by the program.

See Round 2 and Round 1 for more information.

Heraldk

Playoff Predictions, Round 2, 2008

I posted before the first round playoffs started to present some output from my NHL playoff simulator.

How’d it do at predicting the results? Well let’s have a look. I removed the later round predictions from the last post just to clean this up and make it easier to read. Correct predictions are in bold.


Montreal_Canadiens  0.720
Detroit_Red_Wings   0.811
Boston_Bruins       0.280
Nashville_Predators 0.189

Pittsburgh_Penguins 0.590
San_Jose_Sharks     0.634
Ottawa_Senators     0.410 Calgary_Flames      0.366

Washington_Capitals 0.481
Minnesota_Wild      0.468
Philadelphia_Flyers 0.519 Colorado_Avalanche  0.532

New_Jersey_Devils   0.477
Anaheim_Ducks       0.414
New_York_Rangers    0.523 Dallas_Stars        0.586

Not bad huh? In every case, the favourite team of each match was the one that won. Three of the series went right down to the wire though, and I honestly thought Washington was going to beat Philly in game 7. They were a feel good story this year, along with the Oilers late season push to the playoffs that fell *just* short. San Jose and Montreal almost choked in the first round on series leads, but managed to pull through in the end.

So what does the simulator say for the next round? Once again the first column is the probability of winning that round. The second column is the probability of winning the conference final, and the third column is the probability of taking home the cup. This run is one million playoff simulations seeded with the results of the first playoff round.

Montreal_Canadiens  0.616 0.348 0.147
Pittsburgh_Penguins 0.568 0.289 0.114
New_York_Rangers    0.432 0.189 0.064
Philadelphia_Flyers 0.384 0.174 0.056

Detroit_Red_Wings  0.772 0.530 0.380
San_Jose_Sharks    0.483 0.180 0.094
Dallas_Stars       0.517 0.199 0.107
Colorado_Avalanche 0.228 0.091 0.039

Detroit, Dallas, Montreal, and Pittsburgh are the favourites according to the simulator. The only selection that is a bit surprising for this round is Dallas – but I’m not too surprised by that considering some of the numbers they put up this year. If they can shut down Thornton, they’ve got a pretty good shot at the conference finals.

Overall Detroit still has the highest probability of winning the cup, up to 38% now. Montreal and Pittsburgh are neck and neck for second place in probability of winning the cup. Dallas and San Jose are next and the last three have a 6% or less chance.

Now that the first round is out, the last remaining Canadian team is my second favourite team in the league: the Montreal Canadiens. So since the Oil are down and out, I’m cheering for them. Push that 14.7% edge boys!

Heraldk

Predicting the NHL Playoffs 07/08

I’ve become increasingly interested in computer solutions and simulations for domains where you might not expect something can be done. One domain that I’ve been more and more interested in is NHL hockey — a sport that I’m a fan of.

So before the 2007/2008 playoffs started, I decided to write a simulator that figured out how likely it was that each team won the stanley cup based on their regular season statistics. The results are below, with each team’s probability of making it past each round the numbers in the columns. So the Montreal Canadiens have a 72% chance of beating the bruins, and a 12.2% chance of winning the cup.


 Montreal_Canadiens  0.720 0.446 0.264 0.122
 Pittsburgh_Penguins 0.590 0.362 0.193 0.083
 Washington_Capitals 0.481 0.224 0.097 0.033
 New_Jersey_Devils   0.477 0.217 0.096 0.034
 New_York_Rangers    0.523 0.250 0.116 0.043
 Philadelphia_Flyers 0.519 0.236 0.110 0.039
 Ottawa_Senators     0.410 0.167 0.086 0.030
 Boston_Bruins       0.280 0.098 0.039 0.010

 Detroit_Red_Wings   0.811 0.622 0.449 0.330
 San_Jose_Sharks     0.634 0.368 0.150 0.082
 Minnesota_Wild      0.468 0.187 0.056 0.024
 Anaheim_Ducks       0.414 0.182 0.069 0.033
 Dallas_Stars        0.586 0.306 0.141 0.079
 Colorado_Avalanche  0.532 0.178 0.068 0.031
 Calgary_Flames      0.366 0.086 0.037 0.015
 Nashville_Predators 0.189 0.072 0.030 0.012

What’s startling is Detroit’s enormous 33% chance of winning the whole thing. A couple of other surprises include Dallas’ relatively high probability of winning the cup (7.9%). It’s much higher than the Ducks or Avs which are next to them in the standings.

How does this work? I basically calculate the average goals a team is expected to score against their opposing team by using the season goals-for and goals-against numbers. I use this to simulate games between teams and run entire playoff simulations. The probabilities above are simply the number of times that team reaches the next round divided by the number of simulations.

I don’t have very many features added in that I would like to add. One particular one is a recency statistic that increases the relevance of the last 20 games of the season. I expect that such a statistic will greatly increase San Jose and Washington’s chances of winning the cup and probably dampen Dallas’ chances.

Credit to Darse for giving me some hints and tips on how to create my simulator. Hopefully I’ll get a chance to work on it a bit more so that I can build in some more statistics to help it make a more informed decision.

Heraldk

3-0 Baby!

I’ve been playing floor hockey with the CS team again this year. This year we’ve got a pretty competitive looking team. Several new additions to the team mean we’ve got some serious firepower on the floor almost every shift, and the players that don’t have that firepower are good solid players. So with last night’s 10-1 win capping off the 3 game round robin, we’re on top of the division for the first time since I joined the team several years ago. Woohoo!

It’s a lot of fun to play and have a shot to win. I haven’t often been on a team that has had a competitive shot to win.

In related news, it looks like I might be able to join a floor hockey team in the Edmonton ball hockey league. I’ve been looking for a league to play in for a long time, so I’m pretty excited about getting a chance to play finally. The short intramural season was just not enough time to play floor hockey each semester.

Heraldk

Game On

The past three games have been very encouraging for the Oilers. We’re battling through some pretty intense injury trouble, but with last night’s win in LA, we’re now just one game shy of .500, and a pair of points out of a playoff spot.

The two wins against Anaheim are HUGE for a couple of reasons. The first is obviously that wins are important against any team. But beating down Anaheim means that we are more likely to finish above them in the standings – meaning the draft pick we have from them is worth more than the one we gave up for Dustin Penner. The Kevin Lowe vs Brian Burke war seems to be evening out a little more. If the oilers can turn this season around, I think the Dustin Penner offer sheet was a pretty good move for Kevin Lowe – particularly since Penner has been playing better and better over the last few games.

The Edmonton Oilers, in the past couple of years, have been a frustratingly fun team to watch. Even in 05/06 when we made the run to the cup, the Oilers just squeaked into the playoffs. They did so by beating the good teams, and losing to the bad teams … which made no sense. Last year’s season was a terrible disappointment, but the team had so much trouble scoring. Add in the injury troubles that plagued the oilers in the late season, and it really was a recipe for disaster.

I still think the experts are crazy to think they can just write off the oilers as contenders. Anything can happen in the new NHL, and the oilers have a lot of the right ingredients to turn things around. Granted, they have a lot of young and unexperienced players on their team, but there is a lot of skill there, and if the elements click into place, I wouldn’t be surprised to see this team playoff bound.

Heraldk

Maddening Oilers

Man, I just can’t shake the feeling that the Oilers have some chance to go on a stretch of wins. They’ve got so much talent – they just can’t seem to put it together. Of course, the crazy number of injuries isn’t helping. If the Oilers started getting healthy all of a sudden, I think we’d be looking at a particularly good team.

The game last night against Vancouver was a lot of fun to watch. It was also a nail-biter, and every time Vancouver got some pressure on us I got nervous. But Mathieu Garon managed to put on A SHOW and stopped everything that Vancouver sent his way – including three shootout attempts! The Oil looked dangerous for parts of the game, but they seem to have this bad habit of giving the puck away in the neutral zone (or worse, in their own end). I don’t know how to get them to do this, but they need to be able to maintain control of the puck for longer parts of the game or we’re going to continue having trouble maintaining any pressure.

Tonight’s game was worse for that score. We just couldn’t keep constant pressure on the puck and it seemed like every second shift we’d give the puck away several times. Ewww. Hockey like that is maddening to watch for the fan. Glad to see Jarret Stoll get a couple goals though. They were a bit ugly, but I’m glad to see him snap his goal drought. Let’s hope he continues to recover last year’s form!

Maybe their play tonight was due to the back-to-back schedule. I sure hope they can start building some consistency into their game … if they do, and start to get healthy … I seriously think we’ve got a playoff bound team. Until then though, we’re in some trouble.

Heraldk

Catchup

Okay, it seems these catchup posts are getting to be the norm. I hope they aren’t though – so I’m gonna try and get caught up here and start getting myself back in the habit of posting here.

First up, the big thing lately has been my return to poker. I started playing a few weeks ago to catchup on some PSO promos that they’ve been adding with remarkable regularity. I started with Carbon Poker where I jumped right in at 100nl and did quite well and banked a $500 profit plus my $120 PSO bonus. Next was Action poker, which I didn’t do quite as well at. Even with the $100 deposit bonus, I was only able to claw my way back up to even a little after finishing off the $120 PSO bonus. I didn’t really like Action poker – there wasn’t enough tables, and I never really felt comfortable. So I was quick to move on. My next promo though was Cake Poker, and it feels much better. Being up $500 plus the $90 PSO bonus doesn’t hurt in that respect … but I’m also rather enjoying the fast bonus clearing rate. So I think I’m going to stick around at Cake for awhile before I move on to my next PSO bonus.

Work is going well. I just moved offices, and have started to get familiar with the code that produced the bots that played in the man-machine match this past summer. I’m super excited to work on the next iteration of these bots … and even more excited to pit them against the world’s best players. At the moment we’re trying to lock down some money to challenge some players with. Any ideas where we could get that money? We’d ideally like something on the order of $50K or more to use as prize money. I don’t have many ins with people, but if you have any ideas I’d love to hear them.

Diane and I are doing well. We’re currently in the midst of wedding plans which are going pretty smoothly I think. We’ve got several months to go, but we’ve got a lot of the important things locked up and several others in progress. Hopefully we can have invitations out in a month or so. We’ve been pretty busy with social stuff too – attending quite a few fun events over the past month. Too many to really recount in a short space here, but the latest was a Chili cook-off. We made some pretty awesome bison chili (if I do say so myself) along with some skillet cornbread, apple crisp, and cookies. Mmm – it was quite a cooking day! We had a lot of fun in the kitchen, and it’s making me want to force myself to spend more time cooking on a regular basis. I’d certainly eat better (and healthier) if I did.

The Oilers are maddening to watch at times. The most recent game against the flames was sweet to watch though – now if only we could put together an effort like that on a more consistent basis. I see a lot of potential on our club. If we get a few key members back and contributing, we could have a contender on our hands. I realize it’s a lot of ifs, but what else do we have? In particular, I’m really impressed with Robert Nillsson’s play. He’s been fast, creative, and strong with the puck – I’m starting to get excited every time I see him get the puck. One last thing: what the hell is with the hit on Hemsky? That should’ve got a suspension for sure! He could’ve been badly hurt on a play like that!!!!

Not bad for an update post huh? If you’ve got questions, let me know – I’m feeling a bit lonely in blog land!

Heraldk