Questions for Rowsus

Rowsus · 15 Jul 2013

I have had a major crash on my main computer.
This means I will be a lot slower doing my usual things, and have lost a little work.
I have back ups, but it will take me a Little while to bring them totally up to date.
It also Means you might some strange Things in my typing. Like the Capital M in Means and T in Things. Her computer is Danish, and I can only assume it is something to do with her strange keyboard!
I will try and catch up as soon as I can.

SC FANATIC · 16 Jul 2013

Rowsus said:

Once again thanks for the reply Rowsus

.

I actually found out about SCS by the exact same article but never actually made an account until the end of last year. I remember the articles Impromptu wrote being in a completely different area than the forum and was really interested in the strategies he employed to take out the SC title. Was naive in thinking everyone picked their players solely based on durability, point scoring rather than worrying about main competitors etc but this is also due because i have never been in any contention of winning the title. When i eventually did find out about the forum i was amazed of how many serious players with a good knowledge of SC were on the site and time many posters like yourself put into the site and have stuck with the site ever since.

Surprising to me that Bellchambers is that inconsistent never really saw him as that kind of player. And Eric Mackenzie is even more surprising considering he is a dour, defender. But looking at his stats he has had 2 huge scores this year. The others I would put into the mould or being erattic hot and cold players.

P.S I love the way you narrow down the sample size to eliminate misleading data to get an accurate result

Once again cheers for your time

Courtesans · 16 Jul 2013

Hi Rowsus,

I have liked the 30 trades this year, it certainly provides a stimulating challenge, as most of us have found with few trades left and a fair way to go.

I do not like the uneven playing field with loopholes that advantage players with more computer time for various reasons.

What do you think at this stage of the season?

Cheers,

Keith and Coco.

Rowsus · 17 Jul 2013

Courtesans said:

I think they have fallen for an old trap, of trying to satisfy too many people.
Someone once told me, if you try too hard to try and please everybody, you end up pleasing nobody!
The extra trades help keep the non-planners, that used to burn all their trades by round 13, and still have teams that look like a round 1 team, from dropping out of competitions. Less ghost-ships lead to more interest, and longer interest in the competition. Against that, they added a level of difficulty/flexibility that was possibly designed to help separate the chaff from the wheat. The rolling lock-outs. The obvious problem was, it helped a smaller group of people. The people that not only have the access to the internet 24 hours/day, but are willing to commit 6 months of their weekends to the concept. Not everyone has a smart phone, or live/play in areas with immaculate reception, or are prepared to put 18 hours of their weekends into it for near on 6 months.
Like most things, I think there is a compromise available to them.
Removing the Captain loophole would seem like a good start. Lock them in on Fridays game, then you have the true use of what your Vice-Captain was designed for!
In the real AFL, not all late changes are like for like. A Ruckman can be a late out, and be replaced by a small forward. In the real AFL, clubs can't just change players, without good cause, just before a game. Why don't we have a similar system in SC? In the AFL no team has ever taken to the field one player short, why should we? List your 8 non-players from 1 to 8 as to what order they should replace missing players. Some people will say that reduces a skill element, others will argue you are just introducing new loopholes. To counter act this, make a rule, that any players you have in your starting 22 that are not selected for the weekend games, will not be replaced by emergencies! This calls for a commitment from Coaches, that they will look at their teams, and the AFL selected sides, sometime in the 25 hours between selection, and first bounce. Loopholes nearly gone!
There are obvious flaws in this, and it was off the top of my head. I think they need to balance the game more towards "great fun past-time", than near "full time weekend job", if you want to be a chance to win it.

Drew · 17 Jul 2013

^^ Interesting reading Rowsus, really the rolling lockout was pretty much a knee jerk reaction to the Heath Shaw last gasp out, i'd say all the loophole stuff is an unintendined side effect that really just makes it harder for the average player. Think about if you'd just signed up for the first time and an opponent in your league pulls out a loophole trick, it'd feel like someone had cheated! I suspect they might stop it somehow next year.

Bomber18 · 17 Jul 2013

Yea, I'd like the idea of VC and C being locked in after friday.
If your C is a late out, you always have your VC.

IDIG · 17 Jul 2013

Hey Rowsus, how many times a year do you think you could tip the winner and exact margin if you were to tip each and every game of the year? I was talking to a mate and he reckons he's done it twice this year on a Friday night, whereas I haven't done it once lol. Mind you i'm leading one comp by 5, another by 3 and usually get close on a Friday night but yet to strike gold.

Courtesans · 17 Jul 2013

IDIG said:

If I might just chip in here, the trick to winning the footy tips is picking more winners in those 'iffy' games over the year, and that is just luck.

Drew · 17 Jul 2013

Courtesans said:

I remember reading about Swinburne Uni's statistical tipping method a while back. If you are just in a mugs tipping comp at work or at your local pub apparently this is a pretty good way of taking it out.
http://www.swinburne.edu.au/lss/statistics/footytips.html

Rowsus · 18 Jul 2013

Courtesans said:

Spot on, Courtesans. Victory goes to the man that tips the most of those 50/50 games correctly, not to the one that flukes picking an upset now and then.

Rowsus · 18 Jul 2013

IDIG said:

There are a number of different ways to approach this problem, from a mathematical/statistical point of view.
I'll take a simplistic approach, to keep the explanation and maths simple.
Let's say there are 3 easy games, 3 medium games, and 3 hard games each week.
We will allocate a notional points range of 60, 80 and 100 to each of these games.
ie, in an easy game, you might say for example, Fremantle will beat North at Paterson by between 10 and 70. In a tough game, you might say GC could beat Essendon by 30 points, but could lose by 70 points at Metricon. You get the idea. I should point out, by easy game, I am not necessarily talking about it being easy to pick the winner, but easy to narrow the margin down. As an example. You might be looking at Sydney V Geelong at the SCG, and you can be reasonably confident the game will finish between Sydney by 35 and Geelong by 25 points.
Now lets say that most people will pick closer to mean/average of those ranges, when doing there selections, and the result is twice as likely to be within middle part of the points range.
Sounds complicated, but what I am saying is this: Using GC v Ess again.
The results is a 16.66% chance to be GC 6 - 30pts, a 66.66% chance to be GC 1 - 5pts, a draw, and Ess 1 - 45pts, and a 16.66% chance to be Ess 46 - 70pts.
Most people's selections, and most results will occur in that middle 66.66% chance range.
If we take a basic approach, and say each point margin has the same chance of occurring:
then in a tough game the average tipster is a 66.66/50 = 1.333% chance of nailing the margin.
in a medium game the average tipster is a 66.66/40 = 1.667% chance of nailing the margin.
in an easy game the average tipster is a 66.66/30 = 2.222% chance of nailing the margin.
This averages down to the average punter being around a 1.741% chance of picking the winner, and margin, in any randomly selected game.
If you are only selecting margins on a Friday night game, it means you are only trying 23 times/year. The percentages boil down to this:
0/23 margins and winners correct - 66.77% of the seasons you tip in.
1/23 margins and winners correct - 27.21% of the seasons you tip in.
2/23 margins and winners correct - 5.30% of the seasons you tip in.
3/23 margins and winners correct - 0.66% of the seasons you tip in.
So your friend is in the finishing stages of a once in 19 years tipping performance.
Let's put it another way. Ask 20 friends/co-workers how they are going with their margins this year. As we are roughly 70% of the way through the season, you are likely to have 3 or 4 say they nailed one spot on, and your friend with the 2 right. The rest should say they have none.

If you tip a margin in all 198 home and away games for the season, you are likely to pick 3 or 4 margins and winners correctly.
A good way of verifying if my percentages are close to the mark, is to compare them to the odds that are bet on picking margins.
The TAB bet on 2 point margins. Generally around the "popular" margins the TAB bet $26. The TAB usually set markets on exotic betting on somewhere around 130-135%. $26 represents 3.846% of the betting market. As it is a 2 point bet, we can halve that, so that becomes 1.923%. We now need to divide that by the percentage the market is set to. Let's use the average 132.5, so 1.923/1.325 = 1.451%. So the TAB would seem to confirm that my "rough" calculations are within the ball park!

Stroppy · 18 Jul 2013

The lockout is just annoying. If you are around it feels like you are forced to check out the teams. If you are not around and there is a surprise vest or late out you lost out (eg Rowsus).

I much preferred the byes with the best 18 out of 22.
Suggest they go with this for the whole season and reduce the need to pay attention plus the impact of late outs or injuries in the first minute.

IDIG · 18 Jul 2013

Rowsus said:

There are a number of different ways to approach this problem, from a mathematical/statistical point of view.
I'll take a simplistic approach, to keep the explanation and maths simple.
Let's say there are 3 easy games, 3 medium games, and 3 hard games each week.
We will allocate a notional points range of 60, 80 and 100 to each of these games.
ie, in an easy game, you might say for example, Fremantle will beat North at Paterson by between 10 and 70. In a tough game, you might say GC could beat Essendon by 30 points, but could lose by 70 points at Metricon. You get the idea. I should point out, by easy game, I am not necessarily talking about it being easy to pick the winner, but easy to narrow the margin down. As an example. You might be looking at Sydney V Geelong at the SCG, and you can be reasonably confident the game will finish between Sydney by 35 and Geelong by 25 points.
Now lets say that most people will pick closer to mean/average of those ranges, when doing there selections, and the result is twice as likely to be within middle part of the points range.
Sounds complicated, but what I am saying is this: Using GC v Ess again.
The results is a 16.66% chance to be GC 6 - 30pts, a 66.66% chance to be GC 1 - 5pts, a draw, and Ess 1 - 45pts, and a 16.66% chance to be Ess 46 - 70pts.
Most people's selections, and most results will occur in that middle 66.66% chance range.
If we take a basic approach, and say each point margin has the same chance of occurring:
then in a tough game the average tipster is a 66.66/50 = 1.333% chance of nailing the margin.
in a medium game the average tipster is a 66.66/40 = 1.667% chance of nailing the margin.
in an easy game the average tipster is a 66.66/30 = 2.222% chance of nailing the margin.
This averages down to the average punter being around a 1.741% chance of picking the winner, and margin, in any randomly selected game.
If you are only selecting margins on a Friday night game, it means you are only trying 23 times/year. The percentages boil down to this:
0/23 margins and winners correct - 66.77% of the seasons you tip in.
1/23 margins and winners correct - 27.21% of the seasons you tip in.
2/23 margins and winners correct - 5.30% of the seasons you tip in.
3/23 margins and winners correct - 0.66% of the seasons you tip in.
So your friend is in the finishing stages of a once in 19 years tipping performance.
Let's put it another way. Ask 20 friends/co-workers how they are going with their margins this year. As we are roughly 70% of the way through the season, you are likely to have 3 or 4 say they nailed one spot on, and your friend with the 2 right. The rest should say they have none.

If you tip a margin in all 198 home and away games for the season, you are likely to pick 3 or 4 margins and winners correctly.
A good way of verifying if my percentages are close to the mark, is to compare them to the odds that are bet on picking margins.
The TAB bet on 2 point margins. Generally around the "popular" margins the TAB bet $26. The TAB usually set markets on exotic betting on somewhere around 130-135%. $26 represents 3.846% of the betting market. As it is a 2 point bet, we can halve that, so that becomes 1.923%. We now need to divide that by the percentage the market is set to. Let's use the average 132.5, so 1.923/1.325 = 1.451%. So the TAB would seem to confirm that my "rough" calculations are within the ball park!

Haha wow that's an answer..and a half! I especially like where you said 'simplistic' and 'simple'

In that I should offer him pretty generous odds that he won't get another 9 this week and even more generous odds he won't do it twice next year

I'm gonna have a stab at it but for someone to tip the right margin twice in one round would be 0.37%???

(this could be embarrassing!)

Rowsus · 18 Jul 2013

IDIG said:

There are 9 opportunities for someone to get it right. So if you use my figures, which I admit were being a little generous towards your friend, and say it is 1.741% chance you will get any one random right, then to get 2 right in one round is:

(0.01741^2) x (0.98259^7) x 36 = 0.965% of the time you will get 2 out of 9 spot on.

As I said, the 1.741% is being generous to your friend. There really aren't 3 games per week you can be very confident with a 60 point spread!
If you use the TAB's percentage of 1.451, it looks like this:

(0.01451^2) x (0.98549^7) x 36 = 0.684% of the time you will get 2 out of 9 spot on.

Anyone reading this, try a little exercise this week, next week, any week.
In the example I used 3 x 60 point spreads, 3 x 80 point spreads, 3 x 90 point spreads, and stated (later) I was being generous to IDIG's friend.
If you think I'm wrong, and it is easier than that, I challenge to use the 3 x 60 + 3 x 80 + 3 x 100 = 720 points, anyway you like on this weeks 9 games. Not just in 60, 80 and 100 point groups. you may choose to use 50 of your 720 points on the Nth v Carl game tonight, and 120 points on Adel V Geel. Just make the 9 games total 720 points.
I bet not too many of you get all 9 games within your points spread!
In fact, I will start a thread, for those that want to test themselves!

Drew · 18 Jul 2013

Rowsus said:

There are a number of different ways to approach this problem, from a mathematical/statistical point of view.
I'll take a simplistic approach, to keep the explanation and maths simple.
Let's say there are 3 easy games, 3 medium games, and 3 hard games each week.
We will allocate a notional points range of 60, 80 and 100 to each of these games.
ie, in an easy game, you might say for example, Fremantle will beat North at Paterson by between 10 and 70. In a tough game, you might say GC could beat Essendon by 30 points, but could lose by 70 points at Metricon. You get the idea. I should point out, by easy game, I am not necessarily talking about it being easy to pick the winner, but easy to narrow the margin down. As an example. You might be looking at Sydney V Geelong at the SCG, and you can be reasonably confident the game will finish between Sydney by 35 and Geelong by 25 points.
Now lets say that most people will pick closer to mean/average of those ranges, when doing there selections, and the result is twice as likely to be within middle part of the points range.
Sounds complicated, but what I am saying is this: Using GC v Ess again.
The results is a 16.66% chance to be GC 6 - 30pts, a 66.66% chance to be GC 1 - 5pts, a draw, and Ess 1 - 45pts, and a 16.66% chance to be Ess 46 - 70pts.
Most people's selections, and most results will occur in that middle 66.66% chance range.
If we take a basic approach, and say each point margin has the same chance of occurring:
then in a tough game the average tipster is a 66.66/50 = 1.333% chance of nailing the margin.
in a medium game the average tipster is a 66.66/40 = 1.667% chance of nailing the margin.
in an easy game the average tipster is a 66.66/30 = 2.222% chance of nailing the margin.
This averages down to the average punter being around a 1.741% chance of picking the winner, and margin, in any randomly selected game.
If you are only selecting margins on a Friday night game, it means you are only trying 23 times/year. The percentages boil down to this:
0/23 margins and winners correct - 66.77% of the seasons you tip in.
1/23 margins and winners correct - 27.21% of the seasons you tip in.
2/23 margins and winners correct - 5.30% of the seasons you tip in.
3/23 margins and winners correct - 0.66% of the seasons you tip in.
So your friend is in the finishing stages of a once in 19 years tipping performance.
Let's put it another way. Ask 20 friends/co-workers how they are going with their margins this year. As we are roughly 70% of the way through the season, you are likely to have 3 or 4 say they nailed one spot on, and your friend with the 2 right. The rest should say they have none.

If you tip a margin in all 198 home and away games for the season, you are likely to pick 3 or 4 margins and winners correctly.
A good way of verifying if my percentages are close to the mark, is to compare them to the odds that are bet on picking margins.
The TAB bet on 2 point margins. Generally around the "popular" margins the TAB bet $26. The TAB usually set markets on exotic betting on somewhere around 130-135%. $26 represents 3.846% of the betting market. As it is a 2 point bet, we can halve that, so that becomes 1.923%. We now need to divide that by the percentage the market is set to. Let's use the average 132.5, so 1.923/1.325 = 1.451%. So the TAB would seem to confirm that my "rough" calculations are within the ball park!

Interesting reading Rowsus, thanks, after looking at this i was wondering if there's a way to use your easy, medium and hard methodology to bet on draws. On a simple note with 198 games paying a nominal win amount of $51 (seems to be the standard minimum win for a draw on sportsbet) you would need 4 draws a year to turn a profit and probably 5 draws a year to make it worthwhile betting on. I was wondering if you think there's a way to eliminate betting on some games while not greatly reducing your odds of picking all the draws that may occur during a year, for example you could probably eliminate GWS vs Essendon this week.

Zim^zuM · 18 Jul 2013

facing a donut in the fwd line. Trying to weigh up if its worth trading Priddis for J.Roughead vie StevieJ.

Opens up the link, allows me to loophole Crouch and Mitchell (haven't checked if I can due to fixture tbh lol) and assuming makes me 100 points this week

It would mean I wont be trading in Grimes as my D6 and sticking with Terlich. Thoughts?

That or cop the donut or sideways nic nat or buddy. Only 4 trades left

Rowsus · 19 Jul 2013

Drew said:

There have been 14,301 games of AFL/VFL football played, for 151 draws.
This is one every 94.71 games, or 1.06% of games.
Generally there are a lot of games, like your GWS v Ess example, where you can be fairly confident it won't be a draw. Also, a draw is less likely than any other margin under about 75 or 80 points, for logic reasons. If scores are level very late in the game, a team will happily rush a behind to take the lead. If they are 1 point, or more up, they are less likely to do so, as they give up possession. So draws are less likely than other margins.
I started to investigate a system to back draws about 10 or 12 years ago. Generally I arrived at this:
Back games to be a draw when the away team is favourite, and longer than $1.50 to win.
Ignore games on neutral venues, but don't assume that all games between Victorian teams at the MCG and Etihad are neutral. For example, StK used to play only 1 or 2 games/season at the MCG. They could be considered an away team there, anytime they played there. I used to use 5 games as a rule of thumb. ie. If a team plays 5 games at a venue in a season, it can be considered a home, or neutral venue.
I haven't kept track of how this has performed the last 3 or 4 seasons, but it was just in front before then. From memory, it was running at about 15% profit on turnover. There is a huge logistic problem in betting into a system, with a low strike rate, and big odds. You need a big bank to fund it! You might possibly place 200 bets between wins! And if you do, you still haven't re-couped all your losses.
In the "modern era" of football (post 1990) there have been 2 fairly long "droughts" between draws:
13th May 2000, Port Ade drew Geelong at Football Park. It was the 3rd game of round 10 that season.
The next drawn game was -
27th April 2002, St Kilda drew Sydney at Etihad (Docklands). It was game 5 of round 5 that season.
There were 331 games played between those 2 draws.
There was another long "drought" between draws just recently.
16th April 2011, Carlton drew Essendon at MCG, it was game 2 of round 4 that season.
The next drawn game was -
2nd September 2012, Richmond drew Port Ade at MCG, it was game 8 of round 23 that season.
There were 367 games played between those 2 draws.
As you can see, no matter what system you use to narrow the number of games you bet on down, you will need deep pockets to fund a system backing drawn games!

Rowsus · 19 Jul 2013

Zim^zuM said:

Looking at it, it doesn't seem like that good of an idea.
After this week, when SJ is back, if Crouch keeps getting picked, you'll have a packed Mid Bench (Crouch with Mitchell or Rocky/Martin). It seems to be a lot of points sitting on the Mid bench, while Fwd and Def benches only look ok. The link it open up, once you put SJ back down to Fwd line, is only between Martin and Rockliff, and you have to choose between Martin, Rockliff and Mitchell as to who gets benched each week. It also won't help your Fwd line, if NicNat gets told to stop playing early, if WC fall out of contention.
You actually can't loophole Crouch and Mitchell through SJ this week. Crouch and SJ are game 8, and Mitchell is game 9. You could loophole Fyfe and Mitchell, though.
Terlich seems fine as a D6. Last week was his only glitch in a little while, and Grimes would have to have a decent question mark still hanging over his head. Take out Terlich's last game, and his previous 8 averaged 94.4! That's good enough for D6!
Doing a sideways on Buddy or NicNat seems preferable to the Priddis out option. If you think NicNat will play out the season, do Buddy, if you think NicNat will get the tap on the shoulder for early surgery/2014 preparation, do him. There are a lot of good options that won't cost you any cash.
Harvey, Walker, Gunston, Rioli, Cloke, Lewis, Mayne and Henderson, are all possibilities.
Good luck

IDIG · 19 Jul 2013

Wait, did you say Cloke??!

Rowsus · 19 Jul 2013

IDIG said:

LOL, I know! My fingers cramped, I got a shooting pain down my left arm, and it felt like an elephant was sitting on my chest as I wrote his name!

Questions for Rowsus

Rowsus

Statistician

SC FANATIC

Rising Star Winner

Courtesans

Rowsus

Statistician

Drew

Bomber18

Leadership Group

IDIG

Leadership Group

Courtesans

Drew

Rowsus

Statistician

Rowsus

Statistician

Stroppy

Rising Star Winner

IDIG

Leadership Group

Rowsus

Statistician

Drew

Zim^zuM

Rowsus

Statistician

Rowsus

Statistician

IDIG

Leadership Group

Rowsus

Statistician