New/uncommon statistics you'd like to see brought into cricket

[Spikey]

Would be nice to see Keeper's bye conceded.

nah not really. most byes are the result of a wild ball that wasn't called wide, or in the case by some balls by spinners, were pretty much impossible to stop.

 

[benchmark00]

Meh.

Runs per dismissal, for batsmen and for bowlers. Tediously under-rated by sneery "I'm too good for stats" merchants.

Even though it never tells the whole story a batting average or bowling average will almost always tell you far more than any other stat. Because cricket is about scoring more runs than the opposition, and averages go to the heart of the matter for both bowlers and batsmen.

In no way shape or form is that true, and you've just underlined the problem what is wrong with people who use conventional stats.

The line 'because cricket is about scoring more runs than the opposition' sums it up how ignorant it is to say these stats are the best. The team with the most runs does win cricket, but there's more to cricket than just individual scores. It's a team game built upon partnerships and maximising your teams score, not your own. So frustrating that people have such a sheltered view of how cricket is played.

 

[benchmark00]

nah not really. most byes are the result of a wild ball that wasn't called wide, or in the case by some balls by spinners, were pretty much impossible to stop.

Yeah pretty much. Would be more useful to use a +/- for keepers.

 

[dontcloseyoureyes]

Wouldn't mind an equivalent of WAR for cricket though I think it would be particularly hard to implement due to the relative lack of cricket compared to Baseball. Is there a way to check the percentage of someone's runs come from "extra base hits" or whatever as a rating/percentage of runs scored from anything more than ones and twos.

 

[Uppercut]

Ummm because it shows how consistent the batsmen is?

Surely that is obvious. Not sure how someone can possibly not see that.

It tells you a little bit about consistency but average+standard dev tells you a lot more.

 

[KiWiNiNjA]

Wouldn't mind an equivalent of WAR for cricket though I think it would be particularly hard to implement due to the relative lack of cricket compared to Baseball. Is there a way to check the percentage of someone's runs come from "extra base hits" or whatever as a rating/percentage of runs scored from anything more than ones and twos.

Yeah, I'm thinking along these lines too.

Something like baseballs OPS (on base plus slugging) would be interesting. On-Base Percentage would be something along the lines of percentage of balls scored off, where Slugging Percentage would be the bolded part of dcye's post.

 

[Spark]

It tells you a little bit about consistency but average+standard dev tells you a lot more.

Hmmm. Has anyone done this here?

 

[Howe_zat]

Dave Wilson's feature on batting medians is here, ftr.

Thread.

 

[zaremba]

In no way shape or form is that true, and you've just underlined the problem what is wrong with people who use conventional stats.

The line 'because cricket is about scoring more runs than the opposition' sums it up how ignorant it is to say these stats are the best. The team with the most runs does win cricket, but there's more to cricket than just individual scores. It's a team game built upon partnerships and maximising your teams score, not your own. So frustrating that people have such a sheltered view of how cricket is played.

I'll ignore the shriller aspects of that post and just say this. There is a whole constellation of statistics that can be considered to attempt a rounded analysis of a player's contribution and ability. But ultimately the boring old analysis remains: which team's players score the most runs per wicket? And that means that, imperfect though they may be, the batting average and bowling average are the two key statistics.

Compared with batting average a statistic such as, for example, scoring shots per balls faced (however useful that might be as a coaching tool) really is quite unimportant.

 

[Days of Grace]

Not sure if it has been mentioned yet (CBF reading all of your thread, Benchy, you ****ing ****), but I'd like to see a stat for bowlers in ODIs, which breaks down their E/Rs at different stages of the 50-over innings. I.e. the first 10 overs, the middle, and the last 10 overs.

Kyle Mill's figures would be astonishing good for overs 1-10 and beyond dire for the death overs.

 

[Scaly piscine]

I'm not overly familiar with what stats are used on media coverage these days given I'm normally following stuff on Cricinfo/TMS.

Used to quite a reasonable degree already is all your regulation stats applied to left/right arm bowling/batting. Particularly relevant for off-spin/left arm spin. This is a big factor these days and it's something that happens a lot in T20 - where you match up the spinner to take it away from the batsman.

Bowling stats vs. batsmen/tailender etc. In some way you really need to factor in the batsman's average to the runs/wickets taken. The wicket of Kallis say would be worth 4/5 tailenders. Whilst a run is a run whoever scores it. You can categorise this into different phases like new ball, old ball or you could say what is this guy's strike rate and average against tailenders (where you standardise the wickets in the same way as Kallis = 4/5 tailenders, so it's not biased by people who got Chris Martin out half a dozen times) by sticking to batsmen with an average of <15. More complicated would be factoring in say having a low strike rate against a tailender when there's a proper batsman smashing it at the other end - but a categorised strike rate would help to get some of this across.

With things like byes it needs to be averaged over the number of overs - 10 byes in 40 overs is not better than 15 byes in 120 overs. A catches per drop ratio would be interesting, particularly if can be done by fielding position as catches at mid-off are going to be generally easier than those at slip.

 

[morgieb]

For limited overs matches, they should do average x sr (or average x er for bowlers, although that's a bit harder to do....)

Ratio of catches/drops has to be there, ditto bye ratio.

Oh, and FCA

 

[Flametree]

I'd like to see a stat for bowlers in ODIs, which breaks down their E/Rs at different stages of the 50-over innings. I.e. the first 10 overs, the middle, and the last 10 overs.

Kyle Mill's figures would be astonishing good for overs 1-10 and beyond dire for the death overs.

Ditto to this. Seems flawed to compare two bowlers' economy rates when one might regularly bowl at the death or in the power-plays, when another only gets used in the middle overs with 5 on the boundary.

Still subjective, of course. Bowling with the new ball and fielding restrictions is good for bowlers on some wickets against some opponents, and a serious disadvantage on others. Bowling at the death is tough if the opposition are 270-3, but a chance to boost your figures when they are 180-7.


It surprises me that there isn't some agreed measure of wicket-keeper skill. A simple percentage of chances taken would work, though you'd need an arbiter of what constituted a chance, and it can't be applied to historical players.

 

[thierry henry]

Meh.

Runs per dismissal, for batsmen and for bowlers. Tediously under-rated by sneery "I'm too good for stats" merchants.

Even though it never tells the whole story a batting average or bowling average will almost always tell you far more than any other stat. Because cricket is about scoring more runs than the opposition, and averages go to the heart of the matter for both bowlers and batsmen.

Thankyou. Certainly a better indicator than benchmark's eyes ffs.

 

[Debris]

If there are going to be new statistics, one simple one I would like to see is the average runs per wicket for tests a player has played in. I would be interested to see how much it varies between different countries and different eras. You could then compare a players batting or bowling average against this stat to see how far they are from the norm for the tests they have played in.

 

[benchmark00]

I'll ignore the shriller aspects of that post and just say this. There is a whole constellation of statistics that can be considered to attempt a rounded analysis of a player's contribution and ability. But ultimately the boring old analysis remains: which team's players score the most runs per wicket? And that means that, imperfect though they may be, the batting average and bowling average are the two key statistics.

Compared with batting average a statistic such as, for example, scoring shots per balls faced (however useful that might be as a coaching tool) really is quite unimportant.

Strange post.

Since when has the team with the best runs per wicket ratio automatically won the match?

Since when has 54* and 40 been more valuable to the team than two scores of 80's??

Never mind the fact that you've totally missed the point of balls scored off as a statistic. Noone has advocated that replacing average. It's better than strike rate though.

Strange, strange post.

Thankyou. Certainly a better indicator than benchmark's eyes ffs.

Thilan Samaraweera is better than Viv Richards.

MY STATS ARE BETTER THAN YOUR EYE!!!

 

[thierry henry]

Strange post.

Since when has the team with the best runs per wicket ratio automatically won the match?

Since when has 54* and 40 been more valuable to the team than two scores of 80's??

Never mind the fact that you've totally missed the point of balls scored off as a statistic. Noone has advocated that replacing average. It's better than strike rate though.

Strange, strange post.



Thilan Samaraweera is better than Viv Richards.

MY STATS ARE BETTER THAN YOUR EYE!!!

You got beef with Samaraweera?

**** Viv Richards btw

 

[Daemon]

Not sure if it has been mentioned yet (CBF reading all of your thread, Benchy, you ****ing ****), but I'd like to see a stat for bowlers in ODIs, which breaks down their E/Rs at different stages of the 50-over innings. I.e. the first 10 overs, the middle, and the last 10 overs.

Kyle Mill's figures would be astonishing good for overs 1-10 and beyond dire for the death overs.

That's a pretty useful one.

Anderson for example would be poor in the death overs and I think Malinga might just be one of the best in the death of all time.

 

[zaremba]

Strange post.

Since when has the team with the best runs per wicket ratio automatically won the match?

Straw man. Yawn.

 

[benchmark00]

Top response. A++++.