Good we got a rugby thread started. Found some of it quite useful and I liked hearing opions about the games.

Can I ask, just how you go about rating the teams? I assume you use a spreadsheet or something where you factor in certain things relating to each team and the computer comes up with a score. Predicting a draw was pretty amazing. Was that a bit of a one off???

A few hours each day over the last few days I've been putting stats together. Starting working on score prediction today.

Basically I'm combining h2h, scorelines and travelling etc to try to come up with the right balance. I'll test it against a dozen or so results and see what comes up.

btw, last time I ran the hurricanes-blues through it came up with 25-27

Done a few tests (only done half the teams):

Chiefs vs Highlanders:
Mine: 30-24,
Real: 19-7,
Spread was Chiefs +4.5.

Crusaders vs Blues:
Mine: 28-25,
Real: 29-38,
Spread was Crusaders -2.5.

Crusaders vs NSW:
Mine: 31-23,
Real: 19-42,
Spread was Crusaders -8.5.

Blues vs Chiefs:
Mine: 29-23,
Real: 20-27,
Spread was Blues -18.5.

Hurricanes vs Blues:
Mine: 25-26,
Real: 26-26,
Spread was Hurricanes +7.

If I had bet these I'd be 4-1, although you probally wouldn't bet NSW +8.5 if you thought the line should be 8. Only a small sample so I'll see how it does against another 5-10 games.

Sportsnut, you were right about the Blues being rudderless and missing Carlos...a decent first 5 would have led the Blues away from trouble and a solid win...but Aii and the league bloke wanted to give the Hurricanes the game...thought the Hurricanes deserved the win...never gave in.

Did you see the number of turnovers by the canes? They'd march up field quite a few times, only to turn the ball over. The blues would then counter attack every time. This led to 2 of the blues tries. It was a good game to watch though, not great rugby but exciting.

I've fine tuned my model, and it's gone 9-1 betting spreads for 10 previous games this season. 3 of those wouldn't really be a bet as a number of my scores matched the spread (e.g. if you suggest 29-26, you wouldn't bet -2.5 even though you'd win. you beat the spread but it's still a bad bet.)

Course I don't expect it to win most of the time, but I do hope and think it'll be profitable.

Scores it has pumped out for this weekend:

Chiefs vs NSW:
28-31. Spread is right for me.

Crusaders vs Highlanders:
31-24. I'm taking +12 on the highlanders.

Brumbies vs Reds:
35-22. With -8 available on the brumbies, another bet for me.

Chiefs-NSW game is on TV in 1 hour.

Still have to do the stormers-bulls (haven't done the bulls yet). I expect it'll probally be a fairly close one. I like the +4.5 for the stormers, I'll see if my spreadsheet agrees with me.

I like Brumbies-Reds to go over 50.5. I think it'll be a case of brumbies putting on the points and the reds trying to keep up.

Also like the chiefs-nsw to go over 50. I feel my total might be a bit high, my model doesn't take into account injuries and their impact on score (i.e. matty rogers). That said NSW have plenty of weapons so who knows.

I think chiefs-nsw should've gone over 50. 17-32. Thanks to the chiefs missed conversion and refusal to take the penalty goal on the last play. Should've gone over anyway, very sloppy first 60 mins.

Well, the Chiefs win makes it 8/1 this year for my line betting system in S12. Good on you Chiefs!!! Personally, I thought it was the bet of the week, especially as NSW were without Rogers and were playing their 3rd straight game on the road. :smile:

[ This Message was edited by: sportznut on 2004-03-27 18:30 ]

I've reset my own record as 1-0 (fair, since I'm using a system now, and not just off my own thoughts).

It's 1-1 on totals now (should be 2-0 grr).

NSW didn't lose because they didn't have rogers, they lost because they didn't do the basics well, like catching the ball :wink: I think in future I need to factor in the home dog trend. If my system says NSW -3, but the chiefs are a home dog maybe I need to add X amount of points to the chiefs. The chiefs were totally dominant though.