NHL System - Algorithmic

Question

I have created a system in an excel document that has taken me various days. It is a result of the following statistics.

Goaltender GAA
Goaltender SV %
Shots per game
Predicted shots taken before a goal will be scored
Goal Differentials
Goals scored by team in comparison to the NHL average
Goals allowed by team in comparison to the NHL average
Home win % (if home team)
Away win % (if away team)
Past 10 record

Each of these statistics has its own function in the formula. To explain the intricacies of the formula would take hours.

Essentially it is as follows:

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.
A score is then determined for the amount of goals scored compared to league average. A score above 1 indicates a high scoring team. A score below 1 indicates a low scoring team.
A score is then determined for the amount of goals allowed compared to league average. A score above 1 indicates a team that gives up a lot of goals. A score below 1 indicates a team that gives up few goals.
Past 10 games and situational record (I.E. home record for home team, away record for away team) effect the equation at a factor of 10%.
Differential is then factored in as well.

This is a very basic outline of the system and does not even begin to communicate the amount of work that has been put in but thus far this season the system has yet to record a losing night. It predicts winning % for each team as well as the total goals in a given game.

Pandalicious · Answer

I have created a system in an excel document that has taken me various days. It is a result of the following statistics.

Goaltender GAA
Goaltender SV %
Shots per game
Predicted shots taken before a goal will be scored
Goal Differentials
Goals scored by team in comparison to the NHL average
Goals allowed by team in comparison to the NHL average
Home win % (if home team)
Away win % (if away team)
Past 10 record

Each of these statistics has its own function in the formula. To explain the intricacies of the formula would take hours.

Essentially it is as follows:

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.
A score is then determined for the amount of goals scored compared to league average. A score above 1 indicates a high scoring team. A score below 1 indicates a low scoring team.
A score is then determined for the amount of goals allowed compared to league average. A score above 1 indicates a team that gives up a lot of goals. A score below 1 indicates a team that gives up few goals.
Past 10 games and situational record (I.E. home record for home team, away record for away team) effect the equation at a factor of 10%.
Differential is then factored in as well.

This is a very basic outline of the system and does not even begin to communicate the amount of work that has been put in but thus far this season the system has yet to record a losing night. It predicts winning % for each team as well as the total goals in a given game.

oilcountry99 · Answer

Thanks for sharing, good luck

Danrules24 · Answer

Will you be posting the plays here?

Pandalicious · Answer

Sure. I will post every result starting today. Results for today will be posted by 4PM EST

Danrules24 · Answer

Posting results, not posting plays before the start of games?

Danrules24 · Answer

&#xD;&#xA;Disregard my previous post. By posting by 4 EST would be before games played today. When you say "results for today" I&#x27;m assuming that is the results from your number run in excel for that day.&#xD;&#xA;Very interested in this. BOL

Pandalicious · Answer

[Quote: Originally Posted by Danrules24]  &#xD;&#xA;Disregard my previous post. By posting by 4 EST would be before games played today. When you say "results for today" I&#x27;m assuming that is the results from your number run in excel for that day.&#xD;&#xA;Very interested in this. BOL  [/Quote]Exactly. Currently working on them now. They will be up within 30 min. Please check back shortly.

Pandalicious · Answer

The predictions are posted below. Please note that a score of 5 or more in Total Goals Estimated Value indicates a prediction of over. A score of 5 or less indicates under.

--

DET (MRAZEK) @ TOR (REIMER)

Home Predicted Scoring 2.505021614

Away Predicted Scoring 2.758162632

Home Chance of Victory 42.49144512

Away Chance of Victory 57.50855488

Total Goals Estimated Value 5.263184246

CHI (CRAWFORD) @ NJD (SCHNEIDER)

Home Predicted Scoring 2.203568411

Away Predicted Scoring 1.879760262

Home Chance of Victory 73.48044836

Away Chance of Victory 26.51955164

Total Goals Estimated Value 4.083328673

DAL (LEHTONEN) @ CAR (LACK)

Home Predicted Scoring 2.95976612

Away Predicted Scoring 3.670286841

Home Chance of Victory 46.72501285

Away Chance of Victory 53.27498715

Total Goals Estimated Value 6.630052961

Pandalicious · Answer

NYR (LUNDQVIST) @ COL (BERRA)Home Predicted Scoring&#x9;2.364778282Away Predicted Scoring&#x9;2.531851292Home Chance of Victory&#x9;40.64196736Away Chance of Victory&#x9;59.35803264Total Goals Estimated Value&#x9;4.896629575PITT (ZATKOFF) @ EDM (NILSSON)Home Predicted Scoring&#x9;2.366960857Away Predicted Scoring&#x9;3.01847785Home Chance of Victory&#x9;20.27675187Away Chance of Victory&#x9;79.72324813Total Goals Estimated Value&#x9;5.385438707CBJ (McELHINNEY) @ ANA (ANDERSEN)Home Predicted Scoring&#x9;2.392752079Away Predicted Scoring&#x9;2.172041695Home Chance of Victory&#x9;47.71378015Away Chance of Victory&#x9;52.28621985Total Goals Estimated Value&#x9;4.564793773

jenjay23 · Answer

Looks intriguing. BOL in your efforts. What do you figure would be the magic number in "chance of victory" to actually make a play ?

Pandalicious · Answer

That is a great question.The short answer is:It is too difficult to determine. Many factors are in play.The long answer is:When I make my bets I find it important to look at many different things. Although each night, the system as a whole is often correct, I like to minimize potential losses by doing additional investigation.My favorite trends/information to examine are:Recent scores against common opponents.Head-to-head matchup history between the two teams.Whether or not either team is on a back-to-back.Injuries.Extenuating circumstances (I.E. coaching change).The way you need to look at it is simple. If a 73% chance of victory is returned for TEAM A, it means given that matchup you can expect a victory by TEAM A in that particular matchup 73/100 times. Sometimes you will get stifled by the incidences of TEAM B winning at 27%.I&#x27;ve seen many nights that teams predicted in the 50%-59% range all win and teams predicted at 65%&#x2B; don&#x27;t fare as well as anticipated.--I&#x27;m not sure any of the above gives you a suitable answer for your question, but I&#x27;m hopeful that it at least provides some help.--As a post script anecdote I&#x27;d like to add that I&#x27;ve created a system of this sort for all of the following sports.NCAA FootballNHLNBANFLMLBIf enough interest is garnished by this system I will post explanations and results for all of the others.

oilcountry99 · Answer

Nice to see someone attempt some actual handicapping around here

Pandalicious · Answer

I appreciate the compliment.This is my passion. I don&#x27;t work. I gamble. I&#x27;m good at it too. Your kind words are appreciated.

Danrules24 · Answer

I have a question. In the Det @ Tor game, you show Det predicted goals at 2.75 and Tor at 2.50, but home team (Tor) favored to 57.5% of time. How do those numbers correlate?

Pandalicious · Answer

Re-read. You have misread haha.

Danrules24 · Answer

[Quote: Originally Posted by Pandalicious] &#xD;&#xA;&#xD;&#xA;Re-read. You have misread haha. &#xD;&#xA;[/Quote]&#xD;&#xA;My apologies. Too much tequila.

Pandalicious · Answer

[Quote: Originally Posted by Danrules24]  &#xD;&#xA;My apologies. Too much tequila.   [/Quote]No worries haha.

Pandalicious · Answer

DET (MRAZEK) @ TOR (REIMER)DET OVER 5 CHI (CRAWFORD) @ NJD (SCHNEIDER)NJD UNDER 5 DAL (LEHTONEN) @ CAR (LACK)DAL OVER 5.5 NYR (LUNDQVIST) @ COL (BERRA)NYR UNDER 5 PITT (ZATKOFF) @ EDM (NILSSON)PITT OVER 5.5 CBJ (McELHINNEY) @ ANA (ANDERSEN)CBJ UNDER 5.5 --RecordSides: 5-1Totals: 1-5Combined: 6-6Surprisingly NJD was predicted accurately as a fairly large underdog.I&#x27;m fairly certain with the juice that this day is just barely below even.Tomorrow will be better... Good luck all.

Pandalicious · Answer

Prediction for tomorrow early game.VAN (MILLER) @ BUF (JOHNSON)Home Predicted Scoring&#x9;2.377404772Away Predicted Scoring&#x9;3.073772833Home Chance of Victory&#x9;38.06048816Away Chance of Victory&#x9;61.93951184Total Goals Estimated Value&#x9;5.451177605

sports_Network · Answer

sides 5-1 &#xD;&#xA;however, after review, this system appears to lean on sides, totals is an entirely different approach (I don&#x27;t see a formulated concept connecting the mathcup of team goals scored to situational-play per defense/SOG), yet, it&#x27;s only the beginning, and the 1-5 night should be looked at closely if it continues to produce losing results, adjustments are in order. appreciate your effort, and sharing some vg material pertaining to handicapping the NHL

Danrules24 · Answer

I&#x27;m especially interested in games where the dog is predicted to win. A good dog win yesterday on NJ.

sports_Network · Answer

NJ &#x2B;120 home dog WIN

Pandalicious · Answer

[Quote: Originally Posted by sports_Network]  sides 5-1 &#xD;&#xA;however, after review, this system appears to lean on sides, totals is an entirely different approach (I don&#x27;t see a formulated concept connecting the mathcup of team goals scored to situational-play per defense/SOG), yet, it&#x27;s only the beginning, and the 1-5 night should be looked at closely if it continues to produce losing results, adjustments are in order. appreciate your effort, and sharing some vg material pertaining to handicapping the NHL  [/Quote]A 1-5 total night is an anomaly... Typically the totals are better than the sides to be honest. Let&#x27;s see how it fares today. So far we have VAN and OVER 5.

sports_Network · Answer

"Typically, the totals are better than the sides to be honest"&#xD;&#xA;well taken&#xD;&#xA;understanding the anomaly of one nights results....

oilcountry99 · Answer

NJ was A -1.5 reverse puck line win to!!

NHL System - Algorithmic

Hot Forum Topics