All Forums | Systems & Strategies

NHL System - Algorithmic

Participants:

Posts: 158

Pandalicious

Prospect

Joined: Sep, 2015

Posts: 484

Posted: Jan. 7, 2016 - 5:56 PM ET #151

WAS (HOLTBY) @ NYI (HALAK)

Home Predicted Scoring 2.511749143

Away Predicted Scoring 2.707288489

Home Chance of Victory 41.94859674

Away Chance of Victory 58.05140326

Total Goals Estimated Value 5.219037632

FLA (LUONGO) @ OTT (ANDERSON)

Home Predicted Scoring 2.304898823

Away Predicted Scoring 2.632270877

Home Chance of Victory 39.94826221

Away Chance of Victory 60.05173779

Total Goals Estimated Value 4.9371697

PHI (MASON) @ MIN (DUBNYK)

Home Predicted Scoring 2.699386615

Away Predicted Scoring 2.302604803

Home Chance of Victory 63.88287451

Away Chance of Victory 36.11712549

Total Goals Estimated Value 5.001991418

WPG (HELLEBUYCK) @ DAL (LEHTONEN)

Home Predicted Scoring 2.745139988

Away Predicted Scoring 2.817585874

Home Chance of Victory 57.39787577

Away Chance of Victory 42.60212423

Total Goals Estimated Value 5.562725862

AZ (DOMINGUE) @ CAL (RAMO)

Home Predicted Scoring 2.583968497

Away Predicted Scoring 2.618024888

Home Chance of Victory 52.61534843

Away Chance of Victory 47.38465157

Total Goals Estimated Value 5.201993386

confucius1

Prospect

Joined: Jan, 2015

Posts: 102

Posted: Jan. 18, 2016 - 7:20 PM ET #152

wow am i pissed at myself for not finding this earlier on.

WHERE DID YOU GO PAN?!?!?!?!

confucius1

Prospect

Joined: Jan, 2015

Posts: 102

Posted: Jan. 18, 2016 - 7:32 PM ET #153

Avg SOG's to score a goal is a factor of avg SOG's. Once that number is derived it is manipulated by use of the goaltenders SV % to determine the goals that should be allowed with respect to the goalie vs shots factor.

if you dont mind me asking. could you explain this a little further. this stat seems VERY INTERESTING to me.

im guessing something like ((SOG/Games played)*(opponents sv %))/(opponents GAA)

Pandalicious

Prospect

Joined: Sep, 2015

Posts: 484

Posted: Jan. 19, 2016 - 1:26 AM ET #154

Quote Originally Posted by confucius1:

Avg SOG's to score a goal is a factor of avg SOG's. Once that number is derived it is manipulated by use of the goaltenders SV % to determine the goals that should be allowed with respect to the goalie vs shots factor.

if you dont mind me asking. could you explain this a little further. this stat seems VERY INTERESTING to me.

im guessing something like ((SOG/Games played)*(opponents sv %))/(opponents GAA)

Alright, let's make up a random goaltender and a random team. Let's say we have the Flyers @ Capitals

We want to predict the goals that the Flyers will score in this situation.

Let's pretend that the Flyers take an average of 30 shots on the road and score an average of 2.6 goals on the road.

With these numbers you will want to calculate the chance of an individual shot going in. To do this you do 2.6/30. This will provide you with a number in decimal form that essentially indicates a predicted reverse SV % of a home goalie against the Flyers.

0.0866666666666667

Let's say that Holtby (Washington's goalie) has a SV % of 0.92.

The statistic of 0.92 is useless to us, we need his REVERSE SV%. Since save percentage has a maximum of 1 you subtract SV % from 1.

1 - 0.92 = 0.08

At this point you will want to average the predicted reverse SV % against the Flyers on the road as well as Holtby's.

(0.08 + 0.0866666666666667) / 2 = 0.0833333333333333

You will then use the number that you have just received and multiply it by the average road shots of the Flyers.

0.0833333333333333 * 30 = 2.5

This result indicates that you can expect 2.5 goals from the Flyers.

I'm hopeful that none of this was confusing. If you have any questions I can try to help you again.

oilcountry99

Veteran

Joined: Jan, 2010

Posts: 1304

Posted: Jan. 19, 2016 - 9:21 AM ET #155

panda

Thanks for sharing, interesting concept on goal prediction. With your historic results how often does the above method displayed in Post#154 predict the winning team and provide you with an 'edge'

Thanks for your contributions here.

Pandalicious

Prospect

Joined: Sep, 2015

Posts: 484

Posted: Jan. 28, 2016 - 5:35 AM ET #156

Quote Originally Posted by oilcountry99:

panda

Thanks for your contributions here.

That is a difficult question to answer.

A lot more goes into my capping then that calculation. I use a variety of other NHL related statistics in my equation. I.E. (Home scoring compared to league avg, Home allowed scoring compared to league average).

In addition to using statistics I also analyze trends between the two teams. I.E. if the matchups consistently go over 5 or if WAS is 8-1 in its past 9 against PHI. (Continuing with previous analogy).

And as a third factor I watch line movement. For example if WAS opens at a -220 and is bet up to a -235 before game time then WAS is the play.

The way I cap my games is in essence a filtration system.

It begins with the use of the numbers that are calculated.

After that I will compare the results of the calculations to previous trends. If the calculations agree with previous trends I proceed to the next filter.

I check if line movement agrees.

So here is an example of my process (note that these numbers are entirely made up).

The game is PHI @ WAS
My calculations indicate a 62% chance of a WAS victory.
If WAS has won more than half of the matchups in the past year check line movement. If not it is a no play.
If line has been moving consistently towards WAS then bet WAS. If not it is a no play.

Now some of you may be curious how many picks make it through all 3 filters in any given night. The answer to that is if there are 10 games I would likely only have around 2-3 picks. (This includes totals).

Oohnana24

Prospect

Joined: Dec, 2015

Posts: 93

Posted: Feb. 4, 2016 - 11:14 PM ET #157

WOW!!! I just read this whole thread pretty much and Math minds blow me away. I wish I was able to understand algorithms, but I am an English/Arts kind of guy and my brain just doesn't work that way. Lucky of you yours does and you can win some money. Congrats, and I hope win big bucks

Pandalicious

Prospect

Joined: Sep, 2015

Posts: 484

Posted: Feb. 7, 2016 - 2:39 AM ET #158

Quote Originally Posted by Oohnana24:

WOW!!! I just read this whole thread pretty much and Math minds blow me away. I wish I was able to understand algorithms, but I am an English/Arts kind of guy and my brain just doesn't work that way. Lucky of you yours does and you can win some money. Congrats, and I hope win big bucks

Thanks for your support man.

All Forums | Systems & Strategies

NHL System - Algorithmic

Hot Forum Topics