tirri's daily nhl predictions, season 2013-2014

tirri · Post by **tirri** » 30 Sep 2013, 13:18

my oh my, the season begins tomorrow on oct 1

i was pretty lazy this summer in model development because the models worked so well last season. but then i got hit by a modeling bug when the summer was nearing its end and i ended up developing a whole new model family and doing all the modeling steps again. so i now have doubled the amount of models i use daily. it all sounds very fancy but as i suspected, this new family of models wasnt really as good as the one i already used last year so i will use these new ones mostly as back up evidence for the old models and to maybe find some new betting opportunities once in a while.

after 3 years of pretty extensive modeling, im starting to run out of ideas on how to make the models better. i have some new variables i'd like to use but im not sure if the amount of work in collecting them and doing all the analysis from the start would be worth the trouble. so if the models seem to work well enough this season, ill probably use next summer to collect and analyse the data for some other hockey league and transform the models to a new league. the models should work just as well on any league. its good that i started from nhl however because the limits are biggest there

i will post my predictions on several tracking sites but maybe ill have time to copy paste them here too. its just a shame because with placing bets on several different bookies and doing tracking on different sites and actually managing this shit takes more time than running the models and doing actual predictions

the first models wont start running before all teams have played about 12 games minimum in about 3-4 weeks so everyone will have to keep their pants on for a little while longer. this was just a heads up so you can start depositing your hard earned money to some bookies

HAVE A GREAT SEASON FOLLOWING TIRRIS UNPRECEDENTED NHL PREDICTIONS AND TIPS. LOVE LUCK AND LOLLIPOPS EVERYONE

par73 · Post by **par73** » 30 Sep 2013, 19:48

i'll place bets with a few hockey fans to test your model in a different light.

BIG KROK V8 SS · Post by **BIG KROK V8 SS** » 30 Sep 2013, 23:39

Penguins.

Pogue · Post by **Pogue** » 01 Oct 2013, 18:03

BIG KROK V8 SS wrote:Penguins.

LOL shit noob team. Bruins are going all the way this year.

par73 · Post by **par73** » 02 Oct 2013, 02:36

Pogue wrote:
BIG KROK V8 SS wrote:Penguins.
LOL shit noob team. Bruins are going all the way this year.

BIG KROK V8 SS · Post by **BIG KROK V8 SS** » 02 Oct 2013, 05:36

Bruins prefer to go 48 years in between championships, per their history. I've been alive for 4 Penguin Stanley cups. Sup fags.

Pogue · Post by **Pogue** » 02 Oct 2013, 13:17

LOL pens have only ever won 3 cups. Nice try Kirk.

BIG KROK V8 SS · Post by **BIG KROK V8 SS** » 02 Oct 2013, 14:50

I said 4 Stanley Cups. I've watched 4 Pens Stanley cups. They lost 1 of them. Still more than the bruins, and they've been around for decades longer. LOL. LOL JAGR LOL LOLOLOL

Pogue · Post by **Pogue** » 02 Oct 2013, 16:04

Bobby Orr > Lemieux, Jagr, Crosby. GG NO RE.

BIG KROK V8 SS · Post by **BIG KROK V8 SS** » 03 Oct 2013, 05:02

I thought this thread was daily predictions. Where are the daily predictions?

par73 · Post by **par73** » 03 Oct 2013, 07:49

my north-eastern american biased predictions:
welp yall better strap up and start placing bets

bruins are gunna win 50%+ of the games they play this year

its up to you to place a bet on them winning (cuz spreads might have you make nothing)

but bruins are beastly, and home games during regular season = away games during the playoffs

you cannot fuck with boston nation, so just acknowledge it and make moves. make money.

sincerely,
capitalist american- paris in flames

tirri · Post by **tirri** » 03 Oct 2013, 08:26

BIG KROK V8 SS wrote:I thought this thread was daily predictions. Where are the daily predictions?

you will have to use your reading comprehension to the max to get to the bottom of this

toronto seems to continue where it left off last season which is allowing lots of shots but still managing to win. last season toronto was one of the teams i lost most money betting against so i will have to be careful with them. i just dont know how much i want to overrule model recommendations because last season that didnt go too well. however there seem to be 1 or 2 teams every season which seem to be doing too well according to a lot of stats and on the other hand 1 or 2 teams that are doing worse and maybe it would be feasible to take those into account. it will be interesting to see how devils do this year as they were the team i lost most money betting for. i have a feeling they will do much better than most people predict.

BIG KROK V8 SS · Post by **BIG KROK V8 SS** » 03 Oct 2013, 15:53

Canada will never win another stanley cup. those days are long gone.

shr0mbus · Post by **shr0mbus** » 04 Oct 2013, 01:08

good to see u predicting another season cant wait for these ones

BIG KROK V8 SS · Post by **BIG KROK V8 SS** » 04 Oct 2013, 04:03

GG Pens. Incoming 82 game win streak.

tirri · Post by **tirri** » 04 Oct 2013, 06:50

season seems to have started pretty much as expected. devils continue outshooting the opponent but at least in the first game schneider wasnt the solution for their goaltending problems and theres no one to score goals. 1 game is of course too little to draw any conclusions from but i will have to keep in mind that i might have to be careful with devils this season too.

in other news, there were several small rule changes made this summer and all of them should increase the scoring a little. it remains to be seen what the cumulative effect will be. if i was predicting and betting on over/under this would be a real problem. im more interested in predicting the winner and the changes shouldnt really effect this too much. some of my models might be effected a little more than others and theres really no way to correct for the scoring changes, if there even will be any, so i will just have to hope that the overall results wont change too much

par73 · Post by **par73** » 04 Oct 2013, 09:54

tirri how would u analyze toronto's roster/coaching changes from last year to this year?
maybe u would have been winning your bets from last year, if it was this years maple leafs.
they do not look as strong as last year.

NewMutator · Post by **NewMutator** » 15 Oct 2013, 01:28

My body is ready

Myrk · Post by **Myrk** » 15 Oct 2013, 17:18

Any predictions tirri?

BIG KROK V8 SS · Post by **BIG KROK V8 SS** » 16 Oct 2013, 05:19

Crosby another MVP season lock it up.

tirri · Post by **tirri** » 16 Oct 2013, 05:29

predictions start when teams have played about 12 games each so in a couple of weeks

tirri · Post by **tirri** » 16 Oct 2013, 16:00

on a happy side note, i solved the web scraping problem for which i asked help for during the summer. it was actually rather easy to scrape all the thousands of pages like this one (http://www.nhl.com/scores/htmlreports/2 ... 020318.HTM) using only R

too bad i dont have time to build models using the new data now that the season is under way. im disappointed in you data nerds, not one person helped me! and to think how much time i spent making great predictions all of last season!!

tirri · Post by **tirri** » 28 Oct 2013, 15:04

Models are running now although predictions arent available for all matches yet. Nhl has started with no big suprises. The average goals scored per game stands at 5,55 right now which is a little higher than last seasons 5,44. This isnt a huge problem concerning the models but it's worth keeping an eye on. The team's that caused the biggest problems model-wise last season look to be continuing the trend but maybe not as strongly as last year. These teams were mainly toronto and devils. But otherwise I havent noticed any teams performing in ways that would make me feel like the models couldnt be trusted.

Bets for monday, 28th october:
My Tracked Plays
1* - [54] Carolina Hurricanes 2.55 vs Pittsburgh Penguins
These plays are being tracked under the username lunatus at Pick Monitor

All in all with how carolina and pitsburgh have played lately, carolina with these odds would be a pretty good bet. The biggest problem is Peters who at this stage cant be considered even an average back-up. I have no ways of determining how much this affects the predicted probabilities and i dont do any manual corrections. Still worth a bet but only a small one.

Thats all. Minnesota (win in reg 2,6) against Chi and vancouver (win in reg 2,06) look barely profitable but not enough to be worthwhile.

par73 · Post by **par73** » 28 Oct 2013, 19:18

Interesting to see some sort of change (perhaps significant), due to either less intense defense or more intense offense, in the AGPG stat this year compared to last. It is only the beginning of the year however, so it will be interesting to see what that stat looks like a month from now.

When are you going to start betting on the Bruins

tirri · Post by **tirri** » 29 Oct 2013, 14:25

Upcoming Picks
Hockey 10/29 7:05PM Anaheim Ducks vs Philadelphia Flyers
$200 Pick: [3] Anaheim Ducks 1.97
Hockey 10/29 8:05PM Ottawa Senators vs Chicago Blackhawks
$200 Pick: [10] Chicago Blackhawks 1.62
Hockey 10/29 9:35PM Toronto Maple Leafs vs Edmonton Oilers
$100 Pick: [14] Edmonton Oilers 2.05
Hockey 10/29 10:05PM Los Angeles Kings vs Phoenix Coyotes
$200 Pick: [15] Los Angeles Kings 1.95

BIG KROK V8 SS · Post by **BIG KROK V8 SS** » 30 Oct 2013, 05:54

tirri wrote:Models are running now although predictions arent available for all matches yet. Nhl has started with no big suprises. The average goals scored per game stands at 5,55 right now which is a little higher than last seasons 5,44. This isnt a huge problem concerning the models but it's worth keeping an eye on. The team's that caused the biggest problems model-wise last season look to be continuing the trend but maybe not as strongly as last year. These teams were mainly toronto and devils. But otherwise I havent noticed any teams performing in ways that would make me feel like the models couldnt be trusted.

Bets for monday, 28th october:
My Tracked Plays
1* - [54] Carolina Hurricanes 2.55 vs Pittsburgh Penguins
These plays are being tracked under the username lunatus at Pick Monitor

All in all with how carolina and pitsburgh have played lately, carolina with these odds would be a pretty good bet. The biggest problem is Peters who at this stage cant be considered even an average back-up. I have no ways of determining how much this affects the predicted probabilities and i dont do any manual corrections. Still worth a bet but only a small one.

Thats all. Minnesota (win in reg 2,6) against Chi and vancouver (win in reg 2,06) look barely profitable but not enough to be worthwhile.

Great pick retard. PENGUINS.

tirri · Post by **tirri** » 30 Oct 2013, 11:05

Wednesday, 30th October

Slim pickings today with only 4 matches and i feel like at least pit v boston and van v detroit odds are right about where they should be. the models are all over Calgary and last year I learned not to manually meddle with the models too much. there are 2 problems however. First, toronto is really screwing up the models with performances like yesterday, allowing 40 shots against edmonton but still winning with a shutout. This alone wouldnt be reason enough to not follow the predictions but secondly, 2 out of calgarys top 4 scorers are out. its still pretty funny to see toronto as a road favourite against a western team in back to back games. Im itching to bet but will restrain myself, maybe ill check the odds for calgary later.

My Tracked Plays
1* - [57] San Jose Sharks 1.80 vs Los Angeles Kings

BIG KROK V8 SS · Post by **BIG KROK V8 SS** » 31 Oct 2013, 04:52

tirri · Post by **tirri** » 31 Oct 2013, 16:16

My Tracked Plays
2* - [1] Anaheim Ducks 2.25 vs Boston Bruins
2* - [4] New York Rangers 1.51 vs Buffalo Sabres
1* - [5] Nashville Predators 2.45 vs Phoenix Coyotes
These plays are being tracked under the username lunatus at Pick Monitor

tirri · Post by **tirri** » 01 Nov 2013, 20:21

season hasnt started too well but with less than 10 bets anything can happen and im still confident about the model. yesterday was a pretty disappointing day with 2 heavy road underdogs in anaheim and nashville playing in my opinion very well considering the odds they were given and i was pretty happy about those bets.
anaheim got a tying goal scored against them with less than 3 mins and lost in the SO. nashville was leading 3-0 after first period and 4-2 with less than 10 mins remaining but still lost in the SO. great day turned into a pretty bad one in few mins. oh well these will even out during the season, im now 0/3 in overtime and SO games.

todays picks:
My Tracked Plays
1* - [57] New York Islanders 2.37 vs Ottawa Senators
3* - [62] Minnesota Wild 1.80 vs Montreal Canadiens
2* - [63] Colorado Avalanche 1.98 vs Dallas Stars
These plays are being tracked under the username lunatus at Pick Monitor

Pogue · Post by **Pogue** » 02 Nov 2013, 02:31

So how about those Red Sox?

Oh btw Kirk,

Steelers 2-5, Pats 6-2.

Pittsburgh should just stick to making shitty eastern euro food.

BIG KROK V8 SS · Post by **BIG KROK V8 SS** » 02 Nov 2013, 05:04

Pogue wrote:So how about those Red Sox?

Oh btw Kirk,

Steelers 2-5, Pats 6-2.

Pittsburgh should just stick to making shitty eastern euro food.

Red sox go 100 years in between championships. considering their last 2 were flukes, you've got ONE in the last 100 years.

Patriots 0 superbowl wins
Steelers 6

Tom brady dick sucking ability - Over 9000
Shittiness of fenway park - infinity
Shittiness of boston in general - unfathomable.

tirri · Post by **tirri** » 02 Nov 2013, 08:22

nice day as ot/so stats evened out fast.

check out http://www.pickmonitor.com/handicappers#handicappers and sort the list by roi

i might have trouble posting predictions today. saturdays have pretty often been bad days for me and i lose money, 13 matches is a little too much to analyze correctly and i will be busy as it is.

tirri · Post by **tirri** » 02 Nov 2013, 15:27

picks so far today but ive only had time to go over about half of the matches. also all the recommendations now are for really big favorites and low odds which i personally dont like but thats how it goes some days

My Tracked Plays
2* - [5] Anaheim Ducks 1.57 vs Buffalo Sabres
1* - [18] Vancouver Canucks 1.69 vs Toronto Maple Leafs
2* - [24] San Jose Sharks 1.50 vs Phoenix Coyotes
These plays are being tracked under the username lunatus at Pick Monitor

tirri · Post by **tirri** » 02 Nov 2013, 17:31

1* - [26] Los Angeles Kings 1.57 vs Nashville Predators

Pogue · Post by **Pogue** » 02 Nov 2013, 17:42

BIG KROK V8 SS wrote:
Patriots 0 superbowl wins
Steelers 6.

LOL

tirri · Post by **tirri** » 02 Nov 2013, 21:46

and final for the day
1* - [8] New Jersey Devils 1.70 vs Philadelphia Flyers

tirri · Post by **tirri** » 03 Nov 2013, 17:31

My Tracked Plays
2* - [51] Dallas Stars 2.20 vs Ottawa Senators
1* - [56] Minnesota Wild 1.54 vs New Jersey Devils

almost forgot to put these here. ill probably stop posting predictions here soon because all this stuff takes a bit too much time.

bad day yesterday but im sure things will even out. it tells a lot about the season when my road picks anaheim and nashville just lost close road games a few days ago after mostly outplaying their opponets. well now i bet on los angeles and nashville managed to beat them on road after getting outshot about 2:1. same for san jose who put 50 shots on net against phoenix and still managed to lose in SO. oh well, even in retrospective i feel like those were good picks and picks like that will bring money in the long run

shr0mbus · Post by **shr0mbus** » 03 Nov 2013, 20:08

alot of ur upset picks if they lose will only lose by a point, last season aswell.. might be more profitable taking +1.5 handicap in the long run =o or might not be.. who knows worth checking out

tirri · Post by **tirri** » 03 Nov 2013, 21:16

shr0mbus wrote:alot of ur upset picks if they lose will only lose by a point, last season aswell.. might be more profitable taking +1.5 handicap in the long run =o or might not be.. who knows worth checking out

Yeah i havent collected +1,5 odds so it would be hard to backtest when it might be profitable. Im not sure where id get the odds easily enough and id need them for several years. Same goes for -1,5 which is a shame cause its a really good way of getting higher odds for some favourites. I do them once in a while but id feel more comfortable if i had tested it all.

In other news senators just scored at 19:50 in the third. Jebus how many of third period comebacks againts my bets during the first week of betting but at least Dallas won in the shootout. If they had lost i woulda been 0-4 in SOs i think and still probably would have been inclined to think that dallas at 2,2 would have been the correct pick

BIG KROK V8 SS · Post by **BIG KROK V8 SS** » 04 Nov 2013, 01:12

Pogue wrote:
BIG KROK V8 SS wrote:
Patriots 0 superbowl wins
Steelers 6.
LOL

LOL I know it is pretty funny when you think about it. All those years of having a team represent your city and 0 championships to show for it.

Pogue · Post by **Pogue** » 04 Nov 2013, 19:48

idiot plz, your math skills are the equivalent of milk man's myth ones. Steelers haven't been good since the 70's and Bradshaw was second string to the Duck Dynasty guy in college LOL.

tirri · Post by **tirri** » 04 Nov 2013, 20:16

My Tracked Plays
1* - [1] Anaheim Ducks 2.06 vs New York Rangers
1* - [4] Winnipeg Jets 2.30 vs Detroit Red Wings
These plays are being tracked under the username lunatus at Pick Monitor

todays predictions. both are barely profitable so only small bets. also id prefer Winnipeg in regulation but pickmonitor doesnt allow regulation picks. my own bet will be winnipeg in regulation

par73 · Post by **par73** » 04 Nov 2013, 20:32

Tirri if you ever moved to michigan i'm sure you would be able to get into a underground hockey betting culture

BIG KROK V8 SS · Post by **BIG KROK V8 SS** » 05 Nov 2013, 05:21

Pogue wrote:idiot plz, your math skills are the equivalent of milk man's myth ones. Steelers haven't been good since the 70's and Bradshaw was second string to the Duck Dynasty guy in college LOL.

when did the "NEW ENGLAND" team even start? in the 80s? They are too embarrassed to name themselves boston, you just seem to take arbitrary ownership of the most failed and astericked franchise ever. your owner sucks so much dick he's the only reason your team is still in existence. they cheated, it was proven, and they still cheat and get the worst calls in the NFL. calling roughing the receiver penalties at will and pass interference if the receiver calls for it. bunch of fucking pussies and jokes, and have and will never EVER win a superbowl. the whole world knows this, come to terms with it.

tirri · Post by **tirri** » 05 Nov 2013, 20:50

superbly frustrating day today as i only had the time to go over the matches now and pretty much all the edge is gone. there are also several annoyingly difficult to predict toss-ups between bad teams like florida v edmonton and carolina vs philadelphia. i prefer florida and carolina with the odds given now but they dont seem profitable enough to bet.

only small bets today as the odds have pretty much settled where they should. vancouver lit up almost all the indicators from the models which is the first time that has happened this season. but like i learned the hard way last season, it's actually not a place for a big bet but rather means that there is probably something going on the models arent taking into account. still i feel like it's by far the best pick today because i can pretty confidently say that phoenix wont finish this high in the standings by the end of the season and their home record is way too good right now.

My Tracked Plays
1* - [53] New York Islanders 2.17 vs Washington Capitals
1* - [56] Columbus Blue Jackets 1.80 vs Ottawa Senators
1* - [57] St. Louis Blues 1.82 vs Montreal Canadiens
2* - [65] Vancouver Canucks 2.05 vs Phoenix Coyotes
These plays are being tracked under the username lunatus at Pick Monitor

things have been going pretty well lately and this feels just like a 0-4 day with no good indicators and the edge mostly gone with these late analyses, so ill bet pretty lightly

par73 · Post by **par73** » 06 Nov 2013, 01:11

The Pittsburgh Steelers did good work last weekend, giving up a FRANCHISE HISTORY high of 55 points, to Tom Brady and the New England Patriots. Great work!

http://s3.amazonaws.com/br-cdn/temp_ima ... iledtd.gif

http://s3.amazonaws.com/br-cdn/temp_ima ... ronktd.gif

http://s3.amazonaws.com/br-cdn/temp_ima ... lerstd.gif

http://s3.amazonaws.com/br-cdn/temp_ima ... ridley.gif

http://s3.amazonaws.com/br-cdn/temp_ima ... dobson.gif

http://s3.amazonaws.com/br-cdn/temp_ima ... blount.gif

Final Score, PIT 31 - NE 55.

gg krok

Pogue · Post by **Pogue** » 06 Nov 2013, 02:51

Par, that was fantastic

BIG KROK V8 SS · Post by **BIG KROK V8 SS** » 06 Nov 2013, 16:42

Yep and you guys also benefitted from tons of bogus penalties called against us, with no penalties called for us. We played like shit, its a rebuilding year. Congrats on your FIRST win against the Steelers ever. Usually we just destroy your weak little pussy football team.

BIG KROK V8 SS · Post by **BIG KROK V8 SS** » 06 Nov 2013, 16:50