Sorry if I'm spoiling your fun.

There's been one manager who sounded like he's so fed up with the results going against him that he might quit. Another has said he's considering quitting depending on future games. And a third who thinks the coding could be wrong based on what he perceives.

I'm hoping that the stats I'm checking will show that there's nothing wrong with the coding. I admit I moan when things go against me but I always acknowledge when I've been lucky.

However I would never quit because of bad random numbers - like you I'm having fun.  

Sorry Fran.

I was way off. 

Each minute we get the following rolls:

1 to see if there's an injury to your team

11 to see if there's a yellow card (+possible extra for straight red)

1 to see if there's a shot chance

That's 1170 random numbers per game per side

48 teams means 56,160 random numbers per league session.

Plus possible penalties if aggressive players rolled each minute.

And if there's a shot chance: up to 10 to see who gets the shot, roll for offside, then is it on target?, and if so is it a goal?. And the random commentary.

I've put my first12 games in so far: actual chances are about 0.5 per game below expected but the first game of the season was about -5. And there's 3 stalls. The opposition are worse off so far, two of the stall games were below 0 by quite a bit, but still wins.

I think the problem with using ratio/percentages is if you get 0 shots which scores as 0% no matter what the expected is. Or if you get a shot at very low chances it skews it the other way.

As for these 1 in 1000 occurrences, I reckon there could be as many as 10,000 dice rolls per week...

Paul,

Brilliant!

I've started on my games for this season recording expected shots (which is the formula without the /90 - fairly sure this is accurate) and actual chances (offside counts as a chance). Where tactics change I multiply by (time used)/90.

The viewer doesn't show old games so am using GP/DEF/OFF to calculate expected shots. I'm also working out difference and ratio for now - easy to delete the unwanted column.

Only done 4 games so far where both teams were below 100% three times, I stalled in my 4th and my opponents scored 130% in one of their games.

Next season I'm going to keep a check on my own and opponents shot chance percentage : actual chances.

I might even do the whole of my Division - maybe even the whole league -depending on how long it takes each week. Also that's a lot of data.

What I do with the results is unsure as of yet. I don't know much about statistics but did read up about standard deviation.

The advice - if we have an expected shots of 5 and get 6 shots do I record that as 120 (or 1.2) or +1?

If 120, do I ignore the very low shot chance e.g 0.5 but you get 1 shot = 200 or is the +/- idea better?

I can't be certain it's in the current code but in the old code there was a 5% chance of a thwarted attack. This adds to the commentary but also gives the defending player a fraction of a rating. So if you have an 8% chance of a shot then it's 1 to 800 (on a d10000) for a shot and 801 to 1300 a thwarted shot.

I think we play around 860 games a season so 1720 line-ups. I know it's not statistically correct but with that many lineups it's not surprising we get 1 or 2 "1 in a 1000" occurrences.

And yellow cards are checked 11 times as often.

I don't know much about Scottish Football (or English to be fair) so I had a look at the current tables to see which teams are in SESL or not.

There are 42 teams in the top 4 tiers with 39 of those in SESL. The 3 that aren't are Kelty, Spartans and Bonnyrigg - why aren't these in SESL? 

The other 9 teams to make up SESL's 48 are CB (tier 6?), and 4 each from the Highland (tier 5 - BRE, FRA, NAI and DEV) and the Lowland (also tier 5 - ALB, COW, EST and BER) leagues.

Other ex-SESL teams that I remember are BUC and HUN in the Highland league and GRE (now Gretna 2008?) in the Lowland league. I remember ANN replacing GRE and Bill mentions EDI replacing BUC, so COV (who I know is a new team) replaced HUN. Some other teams changed names for season 10 (DEV being one I know from managing them).

Yes I know the SESL Cup referees were randomised this season but despite playing 3 games all passive, we still picked up 5 yellow cards due to 4 strict refs and 1 fair one.

5 of our T11 (actually 5 of our top 6) are a booking away from a suspension and they all play passive in a third of their games.

In the first knockout round it's A vs H, B vs G, C vs F and D vs E so I'll do them in that order.

Group A: CLY play HIB for 1st/2nd and HIB need to win by 4 goals to top the group. AI, in 3rd, will qualify unless they lose to DUU by 9 goals and DUM vs ELG is a draw. ELG need to beat DUM to qualify (with that one exception). 3rd/4th, and playing INV next, is anyone's guess. DUU out.

Group H: INV will finish top.QOS and QP out. ALB, HEA and NAI fighting for places and maybe trying to avoid CLY. ALB (7 points and the best GD) play NAI (6 points) with HEA on 7 points with a better GD than NAI.

Group B: MOT top and need to lose to BRE (2nd with 7 points) by 3 goals for the latter to overtake them. STI in 3rd are also on 7 points and play EDI who are out and will qualify but could be anywhere from 1st to 4th. PT play FRA with FRA needing to win, which could take them to 2nd. PT qualify with a draw but need a win to avoid being 4th.

Group G: ANN and ROS 1st and 2nd on 10 points and likely to remain that way with ANN's suprior GD. CEL and GM 3rd and 4th on 6 points and likely to remain that way although CEL's GD is only 2 better. SM in 5th on 3 points can qualify if they beat CEL by 6 or overtake GM with a 9 GD deficit. DEV out

Group C: RR and CB will finish 1st and 2nd, CB need to beat RR to finish 1st. ALO are 3rd and will finish there unless they lose to eliminated HAM and the other game is a high scoring win. DFA play MON with both on 3 points and MON needing to win. To finish 3rd DFA need to make up 5 goals on ALO and MON 8 goals.

Group of death F: AYR and DEE out with "no hopers" KIL in 4th. They play RAN and need to win by 4 to overtake them. LIV are top and can only finish 2nd with a loss to AYR and a STE win. A STE loss or draw and a KIL vs RAN non-draw drops STE to 3rd.

Group D: COV are out. EF have qualified. Very hard to pick who will finish where. FAL play EF and a win means they are through but they can also qualify with a loss. FOR play COW where a win for them also means they're through and a loss doesn't mean they're out (same GD as FAL). A draw is enough for COW and also for EST who play COV.

Group E: STJ will finish top, PET and ABR eliminated. SRA (7) play BER(7) with ARB (6) playing STJ.

We will finish 3rd or 4th with RAN, LIV and STE one of our likely opponents - 1st, 2nd and 5th in T11.I'm gald I only budgeted for 1 knockout game.

I've looked at a few teams and there's plenty with enough cash and TP to make it competitive. 

We won't be bidding.

Kinda gutted to tie with Livingston for the 5/26 MA and so lose, but it was a last minute increase in the bid from 800k.

And we got a much needed DF anyway and some more CP so still good.

I Googled Law of large numbers.

It would still be nice to know if the coding seems to be correct.

If you toss a coin 10 times and get heads every time the chances of getting heads on the 11th toss is still 1 in 2.