Math with %'s question

**allin1** · 03-07-12, 04:14 PM

I once asked a similar but simpler question regarding probabilities for over/under goals with the stats for each team from soccerstats.com but I never got anywhere. There was a book "probabilities for dummies" or something like that but I never made the time to read it.

**pitman** · 03-07-12, 04:37 PM

I figured I would ask the question on the forum before figuring it out for myself. I'll give it another day or so.
Just trying to figure out how to beat "both teams to score" odds

**Waterstpub87** · 03-07-12, 05:42 PM

56.7%?

**Cookie Monster** · 03-07-12, 06:48 PM

You can substitute the "chance for scoring/allowing a goal" for "chance for winning/losing the game", and you are in a very well known territory. Bill James answered that long time ago, search for the "Log5 rule".

Of course, you muct be sure than your initial percentages are correct vs a neutral team.

**littlezola** · 03-07-12, 09:47 PM

i'm not sure if the bill james formula is relevant to this particular problem, if you work it out using log 5, it suggests a roughly 50% chance of team A scoring against team B, which can't be right.

@OP
If you need help visualising it, take an example of two players taking penalty kicks.

Player A scores with 80% of his penalty kicks. (8/10)
Player B concedes 70% of his penalty kicks. (saves 3/10)
and work it out on paper from there

So.....
Team A= 81% chance to score
Team B= 70% chance to concede

average the two probabilites, so 81+70/200 = 75.5%

for a percentage chance of both teams scoring, it would work like a normal accumulator does.

team a (.75.5) X team b (.62) = 0.4681 both to score.

i think that's right.

**pitman** · 03-07-12, 10:06 PM

thank you for your responses.
.81 X .7 = .567
I do not know if this correct, seems a little low...

I will search the Log5 rule. these initial pecentages would be based on home stats if @ home, Away stats if playing on the road(from a minimum of 12 games each team) never taking in acct teams that they played against.
It would take way too much time to figure out true percentages considering each game vs each team, and whether or not it was a neutral match up.
I don't quite see what you mean ,"chance for scoring/allowing a goal" for "chance for winning/losing the game". Maybe I have to read this Log5 rule first.

**pitman** · 03-07-12, 10:09 PM

thanks, I just read your post LZ. wow, 46.8% for both to score still seems low.

A75.5% and B62% seems right but then 46.8 seems a little low.

**pitman** · 03-07-12, 10:13 PM

well for this instance, the odds wer -105 for Both teams to score and -115 for BT not to score. so maybe that is correct @ 46.8

**littlezola** · 03-07-12, 10:54 PM

forget log5 for this one. it's a useful formula but not relevant to your particular problem, as far as i can tell.

yeah, this 46.8% does seem low, but think of it like you would an accumulator/parlay, because that is really what it is. two seperate events need to occur
1)team a: clean sheet/no clean sheet
2)team b: clean sheet/no clean sheet

Team A 75.5 - @1.32
&
Team B 62 - @1.61
=2.12
Both event's need to occur for you to win your bet, so the bet relfects the 25.5% & 38% chances where one of each will not hit.

if you multiply (0.255 x 0.38) = 0.969 - suggests a scoreless draw every 3-4%, which, subtracting the juice checks out.