Game Predictions – 3/21/2006

With so many teams in the playoff race every night seems to have a bunch of games having very significant playoff implications. Tonight is no different. The biggest might be the NY Islanders vs Montreal but Boston-Atlanta and Toronto-Carolina also have a significance in the eastern conference playoff race, especially if the Islanders beat Montreal. In the west Vancouver and Edmonton play the first game of 3 straight games against each other. Vancouver has an interesting schedule as almost all of their remaining games are against teams they are competing for for the final few playoff spots. Their remaining schedule is: Edmont0n-Edmonton-Edmonton-Los Angeles-Minnesota-Minnesota-Anaheim-Los Angeles-Calgary-Anaheim-San Jose-San Jose-Calgary. To say that their playoff fortunes ( and many others too) are still in their own hands is a very true statement.

Home Team Road Team Predicted Winner Confidence HTOdds
Boston Atlanta Boston Some -118
NY Islanders Montreal NY Islanders Good -142
Philadelphia New Jersey Philadelphia Some -136
Columbus Phoenix Phoenix Some 111
Toronto Carolina Carolina Some -103
Ottawa Pittsburgh Ottawa Strong -290
Detroit Nashville Detroit Strong -195
St. Louis San Jose San Jose Strong 152
Minnesota Calgary Calgary Some 100
Edmonton Vancouver Edmonton Some -122

This article has 27 Comments

  1. Not sure when you pulled the HTOdds for today, but they are well off the current consensus. Ottawa is one of the biggest favorites I’ve seen this season, at a -570. Some places have them as high as -600. The Oilers are 5-0 vs VAN this season, and are a -160 to win tonight. There are also a few more off. Is your source like this every day? The disparity in the Sens line is pretty significant.

  2. Actually, the HTOdds are what I calculate by crunching the numbers and can be used as a betting guide. If you find odds that are better than the HTOdds I listed I’d consider it a good risk/reward bet, if it is substantially worse, I’d consider the road team the better bet. Even though Ottawa is likely to win, I would not bet -600 odds since you are taking some risk for very little reward.

  3. dave, the way the Pred’s are playing your system still has the Wings as a strong play IMO it would be good at best any insight ?

  4. Detroits at home and Nashville played yesterday. Both make a big difference in who wins.

    No, the -290 is not a -1.5 bet. What I am saying is that you are going to bet on an Ottawa ML, you want to get -290 odds or better or else the risk/reward ratio doesn’t make sense. Of course, you are unlikely to get a -290 money line so either bet Pittsburgh or don’t bet at all.

  5. So you can get an idea of what I believe is a fair moneyline. Winning at gambling is all about maximizing the risk/reward ratio. If you are going to take a risk, you want to get rewarded for it. If you aren’t going to get much reward, you don’t want to take much risk. Those numbers are to help you get an idea of what, according to my calculations, is a suitable risk/reward ratio is on each particular game.

  6. Another question. You have the Sharks winning but you have the HT listed at +152. You’re saying you should find the sharks at +152?

    In fact the HT’s are really throwing things off. From a tracking standpoint. It’s pretty hard to track the plays if the actual line you are playing isn’t listed. If someone was looking back, tracking your plays it would look like you had the Sens at -290. Which mind you we know better. But if you waited a few weeks and started to backtrack plays, it really wouldn’t be questioned. Because the Pens…..on the road at the Sens. That could fly.

  7. Ok, I see. It would be the Sharks at -152. Ok. But of course you will never find that line also.

    Hmmmmmm……I just don’t know Dave……Maybe you should go back to the old way of doing it.

  8. I mean, if you were still betting every predicted game under you algorithm you’d be taking Ottawa ML at -570 and risking say $500 to make about 80…doesn’t make sense if you’re trying to maximize risk/reward

  9. Sorry to start all that. But I understand now and it’s a very valuable tool to use in connection with the conventional and public betting lines. Thank you.

  10. Tonights picks on here and Proline:

    PHX @ CLB (BOXED) i think Columbus personally
    ATL @ BOS – Boston
    MTL @ NYI (BOXED) i think NY Islanders personally
    PIT @ OTT – this ones scare me but I have to take Ottawa
    NJ @ PHI – Philly owns NJ in Philly in recent yrs… Philly
    CAR @ TOR- as much as I hate to say it…Carolina
    NAS @ DET – Detroit
    CGY @ MIN – Calgary
    SJ @ STL – San jose
    VAN @ EDM – VAN will win at least 1/3 ..but not tonight..EDM

  11. Good favorites: ATL, COL, TOR, OTT, DET, SJS, CGY
    Home team bet: NYI, PHI, EDM

    Strong favorite – 15 of 24 (63%) win in 60 min.
    Good favorite – 22 of 33 (67%) win or tied in 60 min.
    Home team – 8 of 16 (50%) win in 60 min.

  12. Are you guys actually making money with betting on hockey games??? Or are you just gambling and predicting for fun???

  13. Ultimately you have to do it for fun because it is a risky way to make money. That said, prior to the Olympics I bet $5 on every game as predicted in my prediction table and was up marginally after that time. I started with $100 and at various times I was up by as much as $30 and at other times I was down by $30 but after the 2 1/2 weeks I was a few dollars above my original $100. Certainly not enough to retire on but I didn’t go broke either, which is better than most people would do.

  14. dave:

    good shit – i’m a sucker for system-based selections, accurate record keeping and good money management.

    question about $ management:

    do you bet even amounts per game? it would seem your system is a prime candidate for fixed % of bankroll betting. say 2% of your total roll.

    this would be a dynamic #, recalculated each day and could lead to even greater reward with equal risk.


  15. Personally, I think a system-based method of betting is the only way to go. When you bet based on emotion you are more likely to lose.

    I would bet the same amount on every game and that is what I was doing when I tested it. I was betting $5 per game but you could do 2% as well. The down side is that if you get down it will be harder to catch up because you are making smaller bets, and thus smaller payoffs. Also, if you get up it would be easier to lose money since you would be making bigger bets.

  16. aikmaster, I’m betting for fun. I love a challenge to build an algorithm better than my bookmaker can. I’m betting $1 on each and every single game (60 min only). I’m in the business for 12 days, so far I have won $17 (if EDM must loose or tie today) and there were only 2 days when I lost money, both times on Monday 😀 It is really important to follow established rules and don’t panic on the next day if you loose 9 of 10 games today, for example.

  17. As we all learned tonight… stats have zero to do with NHL outcomes having 8/10 upsets occur tonight. Sometimes emotion does better than stats and trends. Thats a first …personally I cant recall last time there was 10 games or more in one given night and 80% of underdogs winning. Only in the new NHL….gotta love it.

  18. I have to agree with you Mr Hockey about the underdogs winning. I’m comming from Germany and I got interested in betting on NHL games not to long ago….nad I already realized that underdogs do win quite often. This does not happen in soccer to often that underdogs win favourite teams are much more likely to win….hope you will do better with your next predictions

  19. I agree. Underdogs do win in the NHL and win very frequently. But trying to predict when an underdog wins is next to impossible so it just becomes a guessing game. I don’t think you can win money in guessing games so what you want to do is minimize the risk and maximize the reward. You do that by studying the numbers and historic events and extrapolating that to predict likely future outcomes. I’ll put an analytical, numbers based betting system up against a guessing game any day. On a night when you say 8 of 10 games were upsets, my prediction algorithm produced a 50% success rate and two of the games I missed were shootout losses. So, I must have managed to pick some underdogs correctly.

    On the season I am posting a 60+% success rate in my predictions. I challenge any of you to find someone else who makes predictions on every NHL game and has a 60+% success rate.

  20. David i think partly why your success rate is around the 60% is also due to the fact you do predict underdogs once in a while. You kinda have to. I track my success over a season as well and im also hoovering around the 60% mark but I tend to take more risk and guess upsets more frequently. Just for fun. If you stick with betting on every favourite, im sure the success rate would be somwhere similar to that as well. It is a pure guesing game… thats the hard part about it..yet thats what makes it so amusing.

  21. The problem is, who is the favourite? Tonight we have Carolina at Buffalo. Is Carolina the favourite because they have more a marginally better record or is Buffalo the favourite because they are the home team? I am not sure who the average hockey fan would consider the favourite. As you will see in my post of todays game predictions my algorithm considers Buffalo the strong favourite to win.

  22. Your right..who is favourited ever? The Vegas lines that come out I have to give them credit becuase for the most part they are usually accurate in whos the favoured team to win. All im saying is if one were to pick the so called “favouite” i think the success rate would probably be somewhere in the 60%-65% area. Just what i think. The only way I see to make that number of correct guessing to go up is to pick an upset here and there. If thats done, that percentage might get higher but at the same time it might drop as well. Who knows… so im reality, its all just a guessing game.

Comments are closed.