What is the cost of wins in T20?

Illustrating the value for money of a strategist...

It’s often a contrarian view in professional sport, but I believe in accountability. I believe in visible win/loss records, and I believe that there should be more scrutiny over decisions made, but these are two things which I don’t really see a great deal of currently in cricket.

As I mentioned recently, I’m reading ‘How to win the Premier League’ by Ian Graham, who was previously one of the driving forces in analytics at Liverpool. It is an extremely thought-provoking book which I highly recommend, and in the book, the author discusses in depth the decisions which were made (decisions which look good or bad with hindsight, and he gives context).

Graham also discusses the records of teams which he worked for (he was previously at Tottenham also) and how those records improved at the clubs when he worked for them. It’s an interesting discussion because in my own head, I know my own records but haven’t been particularly forward in publicising them, simply due to the fact that no one really does in cricket.

However, if it’s good enough for Ian Graham, it should be good enough for me.

So, here they are - my T20 records when working with teams in competitions with full-time presence with the squad, and some discussion of the records for those teams before and after I was there.

Leicestershire (2022) - 14 matches, won 8 (57%). The highest win rate for the club in the T20 Blast since 2017.

I left the club at the end of that season and in 2023, they won 2/14 matches (14%). In 2021, when I was working at the club remotely, they won 6/14 matches (43%). For those two years combined either side of the one year where I was present at all matches, they won 8/28 matches (29%) - half the win percentage compared to when I was at the club.

Kent (2023) - 14 matches, won 7 (50%). I had one season working in the T20 Blast with Kent, with presence at all matches except for a loss at Glamorgan away which I was unable to attend due to a family reason.

Before I joined in 2022, Kent came bottom of the South Group in 2022 (3 wins in 14 matches), and after I left the club one week before the start of the 2024 Blast, they replicated that finishing position in 2024 (4 wins in 14 matches). So, Kent won 7 matches out of 28 combined (25%), across the season before I joined them and the season after I left. Again, half the win percentage for the team compared to when I was at the club.

Birmingham Phoenix (2021-2024) - 31 completed non-tied matches, won 19 (61%). I have been working with Phoenix since the team was founded in 2019, but the first Hundred tournament didn’t take place until 2021 due to Covid.

Overall - 59 matches, won 34 (58%).

It is extremely rare for a team in a level-budget T20 league to win in excess of 60% of matches - I’ve talked about this before in detail, calling it the 60-40 rule. At Phoenix we have just about achieved this, and are one of the few franchises worldwide who have over the last four seasons. At Leicestershire and Kent, two non-Test match hosting counties, both teams will have had significant budget constraints compared to a number of rival counties, so winning in excess of 50% combined is also a positive.

This brings me on to the cost of wins.

Compared to the cost of signing a lot of players, the cost of hiring a strategist is cheap, and even on a basic level offers great value simply as an evidence-based healthy challenge to decisions. However, it is virtually unheard of for the majority of players to add a notable number of expected wins over the average player in a given tournament - there are some examples of this happening, such as Nicholas Pooran in The Hundred for Northern Superchargers recently, but they are few and far between.

In my view, this is one of the greatest inefficiencies in sports teams. If you could hire a player, that in T20, wins you 58% of matches with them compared to winning 27% without them, teams would be fighting over their signature and paying a significant proportion of their budget to do so.

Even in football, these rationalisations are not close to reality and in cricket, this is even more the case. Teams are willing to spend considerable sums of money on players despite there being more evidence that the value of a high-quality analytics department and good strategists represents a far greater return on investment in terms of expected wins (if you don’t believe me, spend some time looking at the research done in baseball).

It brings me back to the discussion last week on the new PCB mentors (read here). I wrote last week “This is, on the surface, a high cost investment from the PCB which is far from guaranteed to yield any dividends. Instead of the $2.7 on these five mentors, they could spend 20% ($540,000) on establishing the best analytical department in international cricket and save over $2m in the process - a move which would be highly likely to have a higher expected win percentage than hiring these mentors.”

The teams which work this out quicker than the rest will have a huge competitive advantage.

Anyone interested in discussing how I can help their team with strategic management and data-driven analysis can get in touch at sportsanalyticsadvantage@gmail.com.