NRL questions

[Mr J]

Building my NRL system now that the S12 season is basically finished. Will probally have plenty of questions. First one:

Is Stadium Australia (or whatever they call it these days) a neutral venue?

The bulldogs played 6 games there last year during the regular season, should they be considered home games? Even if they are considered a home fixture, is it a REAL 'home game'?

[sportznut]

Mr J,

I think the Bulldogs play half their home games at the Sydney Showground and half at Stadium Australia (Telstra Stadium). While Stadium Australia is definitely their home ground, I wouldn't actually say they have a HUGE home ground advantage there. You are going to be confronted with a few problems like that. In S12 basically EVERY match is played on someone's home ground, but in the NRL and more especially in the AFL, there are quite a few games that are played at venues where the 'home' team doesn't necessarily have a decided advantage.

<font size=-1>[ This Message was edited by: sportznut on 2004-05-10 14:35 ]</font>

[Squirter]

G'day fellas,

The Bulldogs ( at least I think it's the Dogs) even have a home game at the Cake Tin, Wellington NZ as well, but thats over for this season....keep your eye on that for early next season Mr.J

[sportznut]

Yeah, the Bulldogs DO play their home game against the Warriors in Wellington. See Mr J, that's something that certainly doesn't happen in S12.

<font size=-1>[ This Message was edited by: sportznut on 2004-05-10 23:09 ]</font>

[Desi]

And they have won most of their games there which is even more unusual...

[Mr J]

Also doing NFL now. So many stats available :grin:

Found a trend that has gone 287-214 at an average price of 1.91 (-110). So that's about a 10% POT. Good enough to suggest it is probally profitable.

Anyway back to the NRL and AFL for now.

[Mr J]

Well I've done 5 teams now. Only tested 6 games so far and am at 4-2. All 6 picks have been dogs. I'm hoping to do all the previous games this season (around 50-60?), and go 30-20 or so.

It's only a slightly altered version of my S12 system, just hoping that it works :wink:

Also noticed that 3 of my totals have hit the spread at the TAB. Do they do regular totals (ie under/over 40 points) or just these spreads (e.g. 20-30 points, 31-40, 41-50 etc).

[sportznut]

You could probably find some bookies that use over/under but the TABs only seem to use the other method.

[Mr J]

Well since I want to start capping NFL as well I've decided to make a new system. It's not really new, using the same idea's as my S12 just much more detailed. I just like to have a proper system that pumps out scores instead of just following trends.

The way I'm making this system it should be universal, i.e. enter the stats for any footy code and it should pump out scores. I'd need about 3 years of stats on each sport. It's obviously aimed at footy codes (specifically rugby league, rugby union and super 12s), but could work for sports like NHL and NBA. Not sure about soccer or AFL though.

[Desi]

Sheesh Mr J...that must be one hell of a system...you know you could make a lot of money puttimg something like that out in the market (after test driving of course)...Good luck!

[Mr J]

I think I could make more keeping it to myself :wink:

Bascailly it just identifies trends, throws them together and produces a score. To give you an idea of how much work it takes to put together a database, I need results against the spread for 2-3 seasons, scorelines for 3 seasons and as many results as possible for important stats in games (eg turnovers, rucks&mauls, linebreaks etc for Super 12, I'll also need things like tries per match etc). That all has to be copied into nearly 90 different categories (more for american sports). Although in excel this doesn't take nearly as long as it sounds.

God bless excel :wink: it's amazing what you can do with it.

Not sure it's worth spending this much time on it though. I doubt it'd do better than 65% at even money, and the packages here probally perform close to that (with what are probally considerably less complex adn time consuming systems). An extra few % won't justify the amount of time I'd be putting into it. None of this even guarantees it's profitable.

The system itself isn't complicated, just the entering and sorting of all the data.

The other option is to finish off the relatively simple NRL system I was developing yesterday, and if it tests well just use that.

[sportznut]

Geez, that sure sounds like a HUGE amount of work!!!

[Mr J]

Sure will be sportz, but time is something I've had plenty of :wink:

It's not too bad. When you consider how much time the pros put in I think I've got a pretty good deal. Handicappers that are successful over a fair amount of picks each year put in a full working week.

Great thing about using systems is that you just update it every now and then and it spits out the scores. It's much more efficient than normal handicapping because you have already done all of the hard work.

I think between NFL, college FB, NRL, super league, super 12s and AFL I could get close to 1000 bets. If I can adapt it to soccer, nhl, nba or sport with a high volume of games I might get up to 1500. Combine that with the other 2500 bets I currently make (inc a number of packages on this site) and another 1000 I'll soon be pick up, that's ALOT of bets.

[goldmember]

GIDDAY, MR J, the nsw tab has u/o on all nrl games:
There not up just yet. they are usually between 44.5 -- 50.5


<font size=-1>[ This Message was edited by: goldmember on 2004-05-12 14:38 ]</font>

[Mr J]

Ah k thanks for that goldmember. Sportz you wouldn't happen to have written down all the '02-'04 rugby league spreads have you?

[Mr J]

Well I've decided to vary the depth of the system depending on the sport (ie how sharp bookies are at that sport. The sharper the bookies the more detailed the version of the system).

My super 12 system looked at 3 main areas which included home ground advantage and road performance, offence and defence, and head to head. The NRL system will have at least 6 main areas. The NFL system will have 12-18. It'll probally takes me a few days to finish the NRL system. I could use the complex version of the system for S12 and NRL but it is overkill. Keeping it simple also saves me ALOT of time :wink:

[Mr J]

Ahhh now I've gotta write down the spread for '03, '02 and maybe '01 if I can find the spreads for that season somewhere.

Sportz, you sure you haven't recorded the spreads on excel or something?

[jakelee]

Hello Mr J,

You can get this year's spreads from:

http://www.racetab.com.au/results/AFRESULT.HTM


I had all this historical odds data once... when I changed PC's I burned it to CD but lost the CD when I moved house. Yes, I was filthy. I think I will have to pay to get it again.

So... this is what I can remember from my analysis:

- on the TAB the best return you can get is backing the away underdogs. That is, take the away team if they are given the start. This yeilded a return over the seasons of about 98.5% if I recall correctly. Which means you are still losing money of course! But it was the best you could do with a very simple system.

- Low scoring games always had the best odds. When the score is low there are less combinations, and it's harder for the fav to cover the spread. Low scoring games are nearly always when both teams have a lot on the line (like making the top :cool:.

- Field goals are much more prevalent AND the odds for winning by 5 or 7 are quite good (but still not over odds).

- Never take the draw. TAB gives terrible odds for the draw.

- There is no such thing as a sure bet. The TAB makes most money when the fav loses. I'm not sure about the TAB, but real bookies don't actually try to balance the betting on both sides. They sweat like we do on the outcome of games, not just relying on thier vig. They must heavily lay the fav. That's where most of the money is going to be. If the fav wins, they're not so well off, if it loses, its cash time. If the fav is $1.10, then bets of $10,000 are common - no one puts down $2,000 let-alone $10,000 on a $4.50 team. They will hype the fav so that you will accept shorter odds. Look at Newcastle v Dogs on the weekend. Who picked Newcastle?

so... You need to look at the hype surrounding a team. When the papers start saying that a team is championship material - you need to figure out why. It's not always because they are.

Good luck with your system! Let us know how you get on!


JL.

[sportznut]

I'm sorry Mr J.

I have the TAB Sportsbet NRL starts for 2002 and 2003, but only in an ordinary old notebook, you know those things people used to write in back in the old days! :smile:

I'm sorry, but I don't know that you'll be able to get the spreads for 2001.

[Mr J]

Jake does that site give any info earlier than '04?

I've got all the '02 and '04 spreads down..

"on the TAB the best return you can get is backing the away underdogs....This yeilded a return over the seasons of about 98.5%"

A decent system could easily turn away dogs into a profitable trend :wink:

"Never take the draw. TAB gives terrible odds for the draw."

Along with every other sport.

"There is no such thing as a sure bet."

I've been an advantage player for a few years, I already know this :wink:

"but real bookies don't actually try to balance the betting on both sides."

Know that one too. They try to gain value just like we do. Lobsided action...bookies balancing action is the biggest myth around. Also why value is usually found with dogs.

"You need to look at the hype surrounding a team"

Ahhh public opinion. Definately our friend. The bookies can't go out on too much of a limb and have to take into account what the public will think. This creates more value :wink:

BTW, I wouldn't bet at a TAB if my life depended on it. Why pay triple the vig?

[jakelee]

Mr. J,

I will see if I can get earlier spreads.

Might take a few days!

See ya, JL.

[sportznut]

Mr J,

Here you go. On the TAB Sportsbet site, you should be able to find all the Footytab starts for 2002 to 2004.

Simply go to http://www.tabonline.com.au/cgi-bin/main.pl?page=/sports/index.html and click on Footytab, then use Results and Search. That should work.

[Mr J]

Yep that's the site I'm using (you pointed it out to me a few months ago). A bit of a pain flipping through each date and writing down the spreads then copying them into excel. Done '02 and '04 now anyway.

[sportznut]

Okay, I guess if you've done 2 years that way so far, you really may as well keep going with 2003 as well.

I was actually thinking for NRL you might want to try the Footytab starts instead of the Sportsbet starts, because it's quicker (they have all 7 matches on the one page). Also, Footytab has spreads on every single game, unlike Sportsbet which doesn't have line betting on some matches. Most of the spreads are pretty similar for both.

<font size=-1>[ This Message was edited by: sportznut on 2004-05-13 19:44 ]</font>

[Mr J]

Yep that's what I was using.

The spreads for 2 seasons should be enough to test the system, and I have decades worth of other stats.

I will be looking at crowd size to see how much of an influence it can be in a game.

How much difference is there in how refs rule the game (you know how the refs style can be a large influence in rugby union). If there is you might find advantages to exploit props.

I might even look at weather. If the weather girl says there's an 80% chance of raining, then you'd think best bet is to judge to game as if it was going to be wet. Totals would go lower and margins should be smaller. I doubt the TAB goes that far (analysing how weather affects the game).