T20 Blast - Table of Justice
Which teams are currently in false positions?
Ahead of tonight’s matches in the T20 Blast, I am publishing my current Table of Justice for the tournament so far. Regular readers will remember my IPL Table of Justice, which you can read more about here.
My Tables of Justice is built on a private model and show the league position which a team’s performances in a tournament merit and is available for purchase for any league in any format of cricket. Contact sportsanalyticsadvantage@gmail.com
This, and all of my upcoming products, which include a full player valuation model, use learned historical underlying performance data - in this case to align current teams with established drivers of success.
So, for example, if a team is top of my Table of Justice - Somerset so far in this year’s Blast, it’s because their current performances align well with successful teams of the past.
All teams are rated as a percentage - any team with a percentage over 100% means that they are outperforming vital metrics which align with historically successful teams. Conversely, any team with a percentage below 100% means that they are underperforming relative benchmarks set by historically successful teams.
At around the halfway point of this year’s group stages, here is the current table, sorted by average % between batting and bowling:-
According to my Somerset are far and away the outstanding team in the tournament so far, although marginally rank second on points per game to Northants. Somerset’s batting is a long way ahead of every other team, which isn’t entirely a surprise given their stacked top order. It would actually be a surprise if they weren’t towards the top of the table for batting.
Only four teams, Somerset, Northants, Surrey and Glamorgan, have produced above-average levels for both batting and bowling, and are all currently in qualification spots and well set for home quarter finals at this stage.
Middlesex and Warwickshire have strong batting numbers but mediocre bowling numbers, which explains at least in part why Middlesex are towards the bottom of the South Group but higher on the Table of Justice. Being a batting-dominated team increases the volatility of outcome for T20 sides, and Middlesex are a prime example here - it is why I’ve repeatedly said that they needed to swerve the signing of Kane Williamson and invest in a second overseas pacer instead.
On the flip-side, Durham, Lancashire, Sussex and Hampshire have strong bowling numbers but mediocre batting numbers, but this possibly can be at least partly explained by the home venues for these teams not necessarily being conducive to attacking cricket. However, I would argue that it is also driven by Durham, Lancashire and Hampshire making questionable selection decisions which contribute to these batting issues.
This leaves six teams - Worcestershire, Leicestershire, Yorkshire, Kent, Gloucestershire and Essex - who have performed at below the level required for both batting and bowling.
Of these, Leicestershire and Kent haven’t had disastrous on-pitch results, but look rather flattered by having 2+ points per game at this stage - unless they significantly improve during the second half of the season, I would expect their results to mean-revert downwards. Conversely, Glamorgan and Middlesex have a lower points per game than would be expected from their underlying data, and have scope to positively mean-revert for the rest of the tournament.
Available soon will be a full ranking system of players for any league (starting with the IPL), and a little further down the line will be my long-term vision of the holy grail for cricket analytics - a full player pricing model which will be able to show a fair market price (differentiated by league) for any professional cricketer in the world.
Essentially, I will be able to provide the exact market price for any player to any team in any league, in order to help them guard against overpaying for players, as well as to help them hunt for undervalued players.
As mentioned previously, all of my upcoming products use learned historical data to create outputs - giving an independent and unbiased output for cricket teams worldwide, the media and other organisations.
To register early interest in any of these, or to discuss further, please email sportsanalyticsadvantage@gmail.com.