We will run through the entire squad of each team in readiness for the games, so we'll do them in the order that they will play. This is always a good habit in case you run out of time before a season starts. If you only do the first few teams you can finish the rest while watching the opening couple of games.
We'll start each with the captain and then alphabetical order.
For the IPL we need to mark players that are international as this later becomes relevant as to whether they will take the field.
We should also put notes of some kind where the stats may be deceptive. Usually because a player has only played a few games, or only bowls occasionally. In fact players who bowl less than 3 balls a game will have their bowling stats left out, since they most likely won't be bowling anyway.
If you really don't have time, or patience, then it's fine to just do those players most likely to play, and then add any others as and when they get picked.
So let's summarise what we have put on the team stat sheets:
Batting:
Runs = Average rounded off
Balls = Runs / Strike rate as a decimal
Fielding:
Fielding = catches / matches
Bowling:
Runs / 4overs = Economy X 4
Wickets / 4 overs = (Runs / 4 overs) / average
To save space the team stat sheets are at the end of this article.
Once the toss is decided we make a short list of the playing 11. We then do our comparisons. We need a mechanical way to not only pick a winner, but also to give us a win %.
The reason it needs to be mechanical is that we should be able to run the numbers up after a game has been played to see how our system went. This is good because some days we just want to take a break, or go out when the game is on. It's also good to know if we would have won or lost without bias. Anytime somebody gives you tips in hindsight there is always a bias. Why do you think horse betting systems can all tell you about the 50 to 1 shot they had last week? It means nothing.
What we are aiming for is a set of definitive rules to be applied quickly. We have 30 minutes from the announcement of the starting XI to the start of the game. In this time we need to run the numbers, check the odds and make our bets. There should be no rush so we can minimise mistakes. It should also be simple enough we can chat with friends or watch tv while we do it.
Let's have a made up Chennai starting XI in runs order and check how our stats work.
Name 
Runs 
Balls 
Fielding 
Runs/4 overs 
Wkts/4 Overs 
MS Dhoni 
35 
27 
2.5 


S Raina 
34 
24 
0.45 
29 
1.09 
S Badrinth 
30 
26 
0.21 


M Vijay 
27 
21 
0.37 


D Bravo 
25 
20 
0.42 
33 
1.28 
R Jadeja 
23 
27 
0.33 
29 
0.79 
F du Plessis 
22 
19 
0.29 
28 
1.79 
D Bollinger 
20 
17 
0.23 
30 
1.27 
N Kulasekara 
11 
9 
0.36 
31 
1.13 
R Ashwin 
9 
10 
0.21 
26 
1.26 
S Randiv 
8 
10 
0.38 
29 
1.31 
TOTAL 
244 
210 
5.75 


We put them in runs order as a short hand way of getting a batting order. When we do a real game the batting order will be given and we'll use that order.
First we want a batting total for the 120 balls. We have several options here and we need to pick one.
a) We could just take the scores from the openers down until we get to 20 overs.
But this won't take into account the strength of the lower order.
b) We could average the whole team to get an average and then fit that into 20 overs.
But this won't give any weight to the top order who will bat more often.
c) We could do both a) and b) and average them out. This would give weight to the top order while still taking into account the lower order.
STEP 1
Take the total balls and divide it by 120.
210/120=1.75
Then divide the runs by this new number.
244/1.75=139
STEP 2
Start adding up balls until you get close to 120. In our case we reach 118 after Bravo. So we add up all the runs to Bravo and then add the remaining balls on as runs. In this case 2 to make up 120.
Runs up to Bravo +2 balls remaining = 151+2 = 153
STEP 3
Add them together and divide by 2 to get the teams offensive (batting) score:
139+153 = 292
292/2 = 146
For the bowling we will take a simple, in some ways, approach. We will assume 4 bowlers will bowl the full 4 overs and then have 2 others bowl out 2 each. We'll order the bowlers by runs/4ov to avoid an overly complicated system.
We can knock up a quick table of the best 6 bowlers by runs/4ov and then half the runs and wickets of the 5^{th} and 6^{th} bowlers to simulate them only having 2 overs.
Name 
Runs 
Wkts 
R Ashwin 
26 
1.26 
F du Plessis 
28 
1.79 
S Randiv 
29 
1.31 
R Jadeja 
29 
0.79 
S Raina 
15 
0.55 
D Bollinger 
15 
0.64 
TOTAL 
142 
6.34 
We get a defensive score of 6 for 142.
An easy comparison:
149 batting
6 for 142 bowling
But we haven't factored in the fielding yet.
We can simply compare the fielding scores and multiply the result by an arbitrary number. As the scores are going to be averaged out quite well, and we want the fielding to be important, but not the game breaker, we should base it on our Mumbai numbers. Since the only numbers we have are the Mumbai ones, we'll use them. Better to have only a few real numbers than an abundance of imaginary ones. The difference between batting and boling is 7. So we'll make the fielding about half that. 4 is a good number.
A quick scan of the Mumbai and Chennai team stats will give you the impresssion that the wicket stats are much closer than the fielding stats between players. With less variation we'll need a bigger multiplier to amplify their importance. Excuding wicket keepers the fielding stats seem to fall from about 0.2 to about 0.8. Wickets seem to fall between about 1.10 to 1.50. A large difference which is made larger by more player's fielding being added together than wickets. So I'll just go with double, since while I feel the wicket number ought to be triple or more, fielding is more important in my book. So that's a multiplier of 4 for fielding and a multiplier of 8 for wickets.
Just by the use of the phrase "between about" shows I'm being subjective. This means you don't need to take these numbers. Take a look yourself and set your own. You could even punch all the numbers into a spreadsheet and get the true spread. But I'll stick with just how it appears to me.
So who wins?
Let's grab some numbers out of the air for Mumbai just for calculations sake:
152 for batting
5.2 for 148 for bowling
4.6 for fielding
Here's how we work it:
Chennai batting vs Mumbai bowling = (149 + 148) / 2 = Chennai bat 148.5
Mumbai batting vs Chennai bowling = (152 + 142) / 2 = Mumbai bat 147
Chennai field vs. Mumbai field = 5.75 vs 4.6 = Chennai (1.15 x 4) = 4.6
Chennai wickets vs Mumbai wickets = 6.34 vs 5.2 = Chennai (1.14 x 8) = 9.12
Tally it all up and we get Chennai with a score of 15.22
Okay, so we predict a Mumbai win. Not necessarily by 15.22 runs. The 15.22 is actually our index on how sure we are that they will win. If you want to then go ahead and bet $1522 on the event, you could, but better to use a proper staking strategy, and that's what we'll get to.
Previous champions so we can check our work by comparing other teams to this one. If Chennai come up as the weakest we may need to rethink our numbers.
Name 
Runs 
Balls 
Fielding 
Runs/4 overs 
Wkts/4 Overs 
MS Dhoni 
35 
27 
2.5 


S Anirudha 
26 
20 
0.52 


R Ashwin 
9 
10 
0.21 
26 
1.26 
S Badrinth 
30 
26 
0.21 


G Bailey 
28 
21 
0.31 
48 
0 
D Bollinger 
20 
17 
0.23 
30 
1.27 
D Bravo 
25 
20 
0.42 
33 
1.28 
F du Plessis 
22 
19 
0.29 
28 
1.79 
B Hilfenhaus 
4 
6 
0.09 
28 
1.38 
M Hussey 
41 
32 
0.56 


R Jadeja 
23 
27 
0.33 
29 
0.79 
S Jakati 
14 
11 
0.38 
29 
1.15 
J Sharma 
13 
10 
0.3 
31 
1.19 
S Randiv 
8 
10 
0.38 
29 
1.31 
N Kulasekara 
11 
9 
0.36 
31 
1.13 
Y Mahesh 
9 
9 
0.29 
30 
1.43 
A Morkel 
28 
20 
0.22 
32 
1.19 
A Mukund* 
18 
18 
0.53 
23 
2.5 
S Raina 
34 
24 
0.45 
29 
1.09 
W Saha 
26 
20 
0.5 


S Styris 
26 
19 
0.26 
30 
1.09 
S Tyagi 
2 
2 
0.22 
32 
1.07 
K Vasudevadas 
15 
12 
0.79 


G Vignesh** 
18 
15 
0.4 
24 
1.55 
M Vijay 
27 
21 
0.37 


TOTAL 





Numbers in red are deceptive as they are from stats large enough to count, but small enough to be biased.
*A Mukund has only bowled 105 balls. Worth updating his stats as he will be bowling one would think.
**G Vignesh has only bowled 108 balls.
Name 
Runs 
Balls 
Fielding 
Runs/4 overs 
Wkts/4 Overs 
H Singh 
17 
12 
0.29 
26 
1.04 
A Blizzard 
21 
18 
0.41 


Y Chahal 
6 
8 
0.29 
19.8 
1.19 
J Franklin 
33 
26 
0.29 
33 
0.92 
H Gibbs 
26 
21 
0.48 


D Jacobs 
26 
22 
0.47 


M Johnson 
10 
8 
0.13 
28 
1.38 
D Karthik 
26 
20 
0.62 


D Kulkarni 
4 
6 
0.2 
31.4 
1.15 
R Levi 
31 
21 
0.36 


C McKay 
14 
10 
0.12 
32 
1.46 
L Malinga 
9 
8 
0.26 
27 
1.53 
S Marathe** 
45 
31 
0 


A Nechim 
9 
9 
0.04 
30 
1.17 
P Ojha 
5 
5 
0.22 
27 
1.43 
M Patel 
7 
7 
0.22 
29 
1.33 
T Perera 
12 
10 
0.39 
30 
1.33 
R Peterson 
18 
16 
0.16 
28 
1.08 
K Pollard 
29 
18 
0.54 
31 
1.4 
A Rayudu 
26 
22 
0.5 


J Shah* 
14 
12 
0.75 
34 
2.25 
R Sharma 
32 
24 
0.43 
30 
1.11 
R Shukla* 
0 
0 
0.25 
29 
1.83 
A Singh* 
18 
14 
0.43 
24 
0.8 
RP Singh 
5 
6 
0.36 
30 
1.26 
T Suman 
22 
18 
0.23 
31 
1.05 
P Suyal* 


0 
37 
0.57 
A Tare 
21 
19 
0.47 


S Tendulker 
40 
32 
0.38 
32 
0.51 
A Wankhade 
12 
10 
0.2 


S Yadav 
20 
15 
0.36 
24.8 
1.2 
TOTAL 





*Bowled less than 100 balls, so the stats can be used but are not reliable and should be updated every time the player plays.
** S Marathe has only batted 5 innings. We should update his stats each time he bats.
K Yadav and S Nayak have no first class records. For this reason they were left off the table altogether to save space. Should they take the field we will just have to ignore their stats and treat it as a 10 man team until we have some data. It does seem improbable though that with such a large squad these two will be chosen to play anyway.
Posted 4th April 2012 at 8:40am GMT
© Copyright Darren Clifton 2012