Draft 2: Playoffs
-
- Posts: 7035
- Joined: Sat Nov 07, 2009 8:41 am
Re: Draft 2: Playoffs
OK to answer a few questions and to illustrate how simple my system is I'll show you guys how its done with an example to follow in the next post below:
A cricket game is based on 2 major components, batting and bowling (the bowling stats generally include ground fielding misfields and misthrows are counted against the bowler). Therefore what I will do is normalize the batting and bowling and WK performance (ie create a Z-score) of each team into a composite team score. This will allow us to equate between batting performances and bowling performances, because runs and wickets are entirely different units of measure. Hence the z-scores.
The team with the highest composite score wins the game, even if their batting or bowling performance was better than the other teams. Its not only fair, but stays as true to the actual game of cricket as possible.
So just using a team comprised of 2 batsmen and 2 bowlers here's how it would work:
Team A (bat 1, bat 2, bowler 1, bowler 2)
Team B (bat 3, bat 4, bowler 3, bowler 4)
bat 1 random innings is 15 off 25 balls
bat 2 random innings is 63 off 70 balls
bat 3 random innings is 37 off 49 balls
bat 4 random innings is 43 off 45 balls
bowler 1 random innings is 30 average, 3.00 econ
bowler 2 random innings is N/A average, 5.50 econ
bowler 3 random innings is 15 average, 6.00 econ
bowler 4 random innings is 20 average, 4.00 econ
The averages and strike rates/bowling econ of each player would be divided by the global mean statistic of each to get their normalized batting/bowling score. A bowler who doesn't take a wicket will be judged solely on his economy rate in that innings. Those normalized scores would either be added or averaged to get the team total score:
Team A bat score: (15/30)*(60/75) + (63/30)*(90/75) = 0.40 + 2.52 = 2.92
Team A bowl score: (30/30)*(5.00/3.00) + (5.00/5.50) = 1.67 + 0.91 = 2.58
Team Total score : 2.92 + 2.58 = 5.50
Team B bat score: (37/30)*(75.5/75) + (43/30)*(95.56/75) = 1.25 + 1.83 = 3.08
Team B bowl score: (30/15)*(5.00/6.00) + (30/20)*(5.00/4.00) = 1.66 + 1.88 = 3.54
Team B total score: 3.08 + 3.54 = 6.62
Thus Team B would win the head to head matchup by a score of 6.62 to 5.50.
A cricket game is based on 2 major components, batting and bowling (the bowling stats generally include ground fielding misfields and misthrows are counted against the bowler). Therefore what I will do is normalize the batting and bowling and WK performance (ie create a Z-score) of each team into a composite team score. This will allow us to equate between batting performances and bowling performances, because runs and wickets are entirely different units of measure. Hence the z-scores.
The team with the highest composite score wins the game, even if their batting or bowling performance was better than the other teams. Its not only fair, but stays as true to the actual game of cricket as possible.
So just using a team comprised of 2 batsmen and 2 bowlers here's how it would work:
Team A (bat 1, bat 2, bowler 1, bowler 2)
Team B (bat 3, bat 4, bowler 3, bowler 4)
bat 1 random innings is 15 off 25 balls
bat 2 random innings is 63 off 70 balls
bat 3 random innings is 37 off 49 balls
bat 4 random innings is 43 off 45 balls
bowler 1 random innings is 30 average, 3.00 econ
bowler 2 random innings is N/A average, 5.50 econ
bowler 3 random innings is 15 average, 6.00 econ
bowler 4 random innings is 20 average, 4.00 econ
The averages and strike rates/bowling econ of each player would be divided by the global mean statistic of each to get their normalized batting/bowling score. A bowler who doesn't take a wicket will be judged solely on his economy rate in that innings. Those normalized scores would either be added or averaged to get the team total score:
Team A bat score: (15/30)*(60/75) + (63/30)*(90/75) = 0.40 + 2.52 = 2.92
Team A bowl score: (30/30)*(5.00/3.00) + (5.00/5.50) = 1.67 + 0.91 = 2.58
Team Total score : 2.92 + 2.58 = 5.50
Team B bat score: (37/30)*(75.5/75) + (43/30)*(95.56/75) = 1.25 + 1.83 = 3.08
Team B bowl score: (30/15)*(5.00/6.00) + (30/20)*(5.00/4.00) = 1.66 + 1.88 = 3.54
Team B total score: 3.08 + 3.54 = 6.62
Thus Team B would win the head to head matchup by a score of 6.62 to 5.50.
https://www.youtube.com/watch?v=JjtuZBykSzM (Noreaga - Blood Money Part 3)
-
- Posts: 7035
- Joined: Sat Nov 07, 2009 8:41 am
Re: Draft 2: Playoffs
As of right now, seamers and spinners will be scored separately.
Similarly, batsmen will be scored in the following categories:
openers (1-2)
top order (3-5)
finishers (6-7)
tailenders (8-11)
The stats for openers and top order are almost identical so shouldn't affect people too much. Openers average 27.73 at 76.22 strike rate, and top orders average 26.71 at 75 strike rate. These are the median values of the list that I compiled on page 2 of the Draft 2 thread.
Similarly, batsmen will be scored in the following categories:
openers (1-2)
top order (3-5)
finishers (6-7)
tailenders (8-11)
The stats for openers and top order are almost identical so shouldn't affect people too much. Openers average 27.73 at 76.22 strike rate, and top orders average 26.71 at 75 strike rate. These are the median values of the list that I compiled on page 2 of the Draft 2 thread.
https://www.youtube.com/watch?v=JjtuZBykSzM (Noreaga - Blood Money Part 3)
Re: Draft 2: Playoffs
What if a player was unbeaten for 5 runs in that random innings?
- CrimsonAvenger
- Posts: 9838
- Joined: Thu Dec 13, 2007 2:57 am
- Supports: Mountaineers
- Location: India
Re: Draft 2: Playoffs
Some nice loopholes being brought out there by grant .
So you have one in 35 chance of getting 194* by Coventry if you pick him in your side. Not a bad pick if you are left with some lowly picks towards the end of the draft...
So you have one in 35 chance of getting 194* by Coventry if you pick him in your side. Not a bad pick if you are left with some lowly picks towards the end of the draft...
Re: Draft 2: Playoffs
I'm trying to work it out but maths was never my strong point.
Is 30 the number of innings of Bat 1 and why is it multiplied by 60/75
Is 30 the number of innings of Bat 1 and why is it multiplied by 60/75
Re: Draft 2: Playoffs
I think 30 is the average (or median? Whatever Kriterion_BD chose) score at that batting position and 75 is the average strike rate. Bat 1 scored 15 runs at a strike rate of 60, so that's where the 60 came from.
-
- Posts: 7035
- Joined: Sat Nov 07, 2009 8:41 am
Re: Draft 2: Playoffs
But you'd be short of a bowler unless you have some good all-rounders of which there are very few. Not a good tradeoff for a 3% chance of landing a 194*.CrimsonAvenger wrote: ↑Thu Apr 16, 2020 5:02 amSome nice loopholes being brought out there by grant .
So you have one in 35 chance of getting 194* by Coventry if you pick him in your side. Not a bad pick if you are left with some lowly picks towards the end of the draft...
https://www.youtube.com/watch?v=JjtuZBykSzM (Noreaga - Blood Money Part 3)
-
- Posts: 7035
- Joined: Sat Nov 07, 2009 8:41 am
Re: Draft 2: Playoffs
In the example above 30 is the made-up median batting average. The actual values are 27.73, 26.71, 19.50, and 9.00 for opening batsmen, top orders (3-5), finishers (6 and 7), and tailenders (8-11) respectively.
60/75 is the normalization of strike rate.
https://www.youtube.com/watch?v=JjtuZBykSzM (Noreaga - Blood Money Part 3)
Re: Draft 2: Playoffs
Still not completely understanding, but if the others do and agree to it, sounds good to me.Kriterion_BD wrote: ↑Thu Apr 16, 2020 5:34 amIn the example above 30 is the made-up median batting average. The actual values are 27.73, 26.71, 19.50, and 9.00 for opening batsmen, top orders (3-5), finishers (6 and 7), and tailenders (8-11) respectively.
60/75 is the normalization of strike rate.
Re: Draft 2: Playoffs
There needs to be a way to reward players for longevity. Otherwise one could just pick guys that played a few games and did well.
Neil Johnson, Alistair Campbell, Murray Goodwin, Andy Flower (w), Grant Flower, Dave Houghton, Guy Whittall, Heath Streak (c), Andy Blignaut, Ray Price, Eddo Brandes