Is the Points Tally in IPL Fair?

Cricket
IPL
Author

Anvit Aggarwal

Published

June 1, 2025

Two new teams, namely, Gujarat Titans and Lucknow Supergiants, were added to the IPL in 2022. Since then, the seasons have had 70-match seasons, which means that each team plays only 14 matches instead of the 18 it should have played if the tournament were a fair tournament. By a fair tournament, I mean that all teams would face each other twice, once at home and once away. Currently, the schedule of the IPL results in teams having different levels of opposition. Some teams have a harder schedule while other teams have an easier schedule. I think that this problem is solvable. I used a chess-like Elo-based algorithm to analyse the actual strengths of the teams. All teams were assigned an equal rating of 1200 at the beginning, and I ran the tournament and updated the ratings 1000 times to ensure that the ratings of each team converged to the level of their performance. You can view the graphs that show the ratings of each team after each iteration. Then, I predicted the number of wins of each team from the ratings. In each case, the number of predicted wins matches the actual number of wins. Then, I predicted what would have happened if the tournament were fair. You can view the code here:
Link to Code

Let’s go season by season:

2022:

Points Table:

Ratings:
GT: 1355
LSG: 1306
RR: 1299
RCB: 1254
DC: 1200
PBKS: 1200
KKR: 1150
SRH: 1149
CSK: 1041
MI: 1041

These are the ratings I obtained. We can see that GT is a distant first. Though we have a surprising result here. RR, which came second in the points table and ended as the runner-up in the competition, has a lower rating than LSG, which ended third on the points table and lost to RCB in the eliminator. To ensure that the ratings are accurate, I tried to reproduce the points table again using these ratings.

Expected wins based on Current Ratings:
GT: 10.00
RR: 9.00
LSG: 9.00
RCB: 8.00
DC: 7.00
PBKS: 7.00
KKR: 6.00
SRH: 6.00
CSK: 4.00
MI: 4.00

This matches the actual number of wins for each team in the tournament.
If the tournament was a fair one, as I have defined earlier, the expected points of the teams would have been the following:

Fair Tournament Expected Points:
GT: 25.8 | RR: 23.2
LSG: 23.5 | RCB: 20.8
DC: 18.0 | PBKS: 18.0
KKR: 15.3 | SRH: 15.3
CSK: 10.0 | MI: 10.0

2023:

Points table:

The case is more interesting in 2023.
GT: 1353
LSG: 1277
CSK: 1272
MI: 1251
RR: 1201
RCB: 1199
KKR: 1152
PBKS: 1150
DC: 1100
SRH: 1044

Expected Wins Based on Current Ratings:
GT: 10.00
CSK: 8.00
LSG: 8.00
MI: 8.00
RR: 7.00
RCB: 7.00
KKR: 6.00
PBKS: 6.00
DC: 5.00
SRH: 4.00
Again, matches the actual number of wins. Thus, validating the ratings.

Fair Tournament Expected Points:
GT: 25.9 | LSG: 22.1
CSK: 21.9 | MI: 20.8
RR: 18.1 | RCB: 17.9
KKR: 15.4 | PBKS: 15.3
DC: 12.7 | SRH: 10.0

Surprisingly, LSG again comes out ahead of CSK (the team that went on to win the IPL season that year).

2024:

Points Table

Ratings:
KKR: 1378
RR: 1263
SRH: 1262
LSG: 1206
RCB: 1203
DC: 1202
CSK: 1193
GT: 1124
MI: 1066
PBKS: 1104

Expected Points Based on Current Ratings:
KKR: 9.00
SRH: 8.00
RR: 8.00
RCB: 7.00
CSK: 7.00
DC: 7.00
LSG: 7.00
GT: 5.00
PBKS: 5.00
MI: 4.00

Fair Tournament Expected Points:
KKR: 27.0 | RR: 21.4
SRH: 21.4 | LSG: 18.4
RCB: 18.2 | DC: 18.1
CSK: 17.6 | GT: 13.9
MI: 11.0 | PBKS: 12.9

RR ends slightly ahead of SRH, but the difference is extremely small and can safely be neglected. Here as well, it is surprising to see LSG ahead of RCB.

We can safely say that LSG was pretty unlucky with its schedule in its first three seasons.

2025:

Points table:

Ratings:
RCB: 1333
PBKS: 1327
GT: 1294
MI: 1246
DC: 1245
LSG: 1171
SRH: 1176
KKR: 1116
RR: 1048
CSK: 1045

Expected Points Based on Current Ratings:
PBKS: 9.00
RCB: 9.00
GT: 9.00
MI: 8.00
DC: 7.00
SRH: 6.00
LSG: 6.00
KKR: 5.00
RR: 4.00
CSK: 4.00

Fair Tournament Expected Points:
RCB: 24.8 | PBKS: 24.5
GT: 22.9 | MI: 20.4
DC: 20.4 | LSG: 16.4
SRH: 16.7 | KKR: 13.6
RR: 10.3 | CSK: 10.2

In this year’s IPL, DC would probably consider itself unlucky. Its performance was almost as good as MI but MI edged out ahead. One drawback of this algorithm is that it doesn’t take into account the margin of victory. Given MI’s significantly higher NRR, we could say that MI is probably still a significantly better team than DC.

I would like to explore the usefulness of a similar algorithm for the World Test Championship. Here, no weightage has been given to home and away matches. In WTC, perhaps, the distinction matters. For example, defeating India away is probably significantly easier than defeating India at home. In IPL, I assume that there isn’t much of a difference.