• Welcome to the Cricket Web forums, one of the biggest forums in the world dedicated to cricket.

    You are currently viewing our boards as a guest which gives you limited access to view most discussions and access our other features. By joining our free community you will have access to post topics, respond to polls, upload content and access many other special features. Registration is fast, simple and absolutely free so please, join the Cricket Web community today!

    If you have any problems with the registration process or your account login, please contact us.

Equivalent batting and bowling averages

Howe_zat

Audio File
So I was wondering, what's the bowling equivalent of averaging 40 with the bat? Or the batting equivalent of averaging 30 with the ball? How can you say if someone's a batting or bowling allrounder if they have averages in 30s?

(Disclaimer: This is a thread in which we play with numbers. If you are the sort of person who delights in telling other people that averages aren't everything, as if they don't already know that, the back button is over there.)

I thought the best way to do it was to use the average runs per wicket across all Test cricket in the 21st century - it's designed as an analysis of modern players - averages are a bit higher than they used to be as I'm sure you're aware - and find a way of showing how much better a player is than that 'standard' average.

At the time of writing, that's 32.18. So a batsman who can average twice as much as that should be twice as good as the average batsman, and a bowler who can pay half that for his wickets should be twice as good as the average bowler.

The table below gives a list of equivalent averages for batsmen and bowlers based on how many times better than that they are (given as 'index').

Code:
Batting	Bowling	Index
64.36	16.09		2
61.142	16.93684211	1.9
57.924	17.87777778	1.8
54.706	18.92941176	1.7
51.488	20.1125		1.6
48.27	21.45333333	1.5
45.052	22.98571429	1.4
41.834	24.75384615	1.3
38.616	26.81666667	1.2
35.398	29.25454545	1.1
32.18	32.18		1
28.962	32.50505051	0.9
25.744	36.56818182	0.8
22.526	41.79220779	0.7
19.308	48.75757576	0.6
16.09	58.50909091	0.5
This is clearly a load of horlicks. There's no way that a bowler who could average 21 in the modern game isn't way out ahead of your common or garden 48 averaging batsman.

The flaw in this is that to be a batsman you need to average considerably more than the overall average runs per wicket as you need to make up for your tailenders, but the vast majority of wickets taken are by front line bowlers. So if all your bowlers take their wickets at 32 you'll have a bang average attack that bowls teams out for 320. But if all your batsmen average 32 you won't get to 320 on average because you've also got your tailenders falling short.

So I re jigged it with 38.46 as the batting benchmark - the average rpw in modern Test cricket for a top 7 player. The list goes

Code:
Batting	Bowling		Index
76.92	16.09		2
73.074	16.93684211	1.9
69.228	17.87777778	1.8
65.382	18.92941176	1.7
61.536	20.1125		1.6
57.69	21.45333333	1.5
53.844	22.98571429	1.4
49.998	24.75384615	1.3
46.152	26.81666667	1.2
42.306	29.25454545	1.1
38.46	32.18		1
34.614	32.50505051	0.9
30.768	36.56818182	0.8
26.922	41.79220779	0.7
23.076	48.75757576	0.6
19.23	58.50909091	0.5
Which I like.

A few points

- The best averaging bowler of modern cricket (McGrath) has a roughly equivalent record to the best averaging batsman (Sangakkara), with 1.5 times the output of an average bowler and batsman respectively

- A bowling equivalent of Don Bradman would average 12.38

- Most allrounders are closer to an average bowler than an average batsman - Flintoff for example had the record of 1 average bowler but about 0.82 average batsmen. This makes sense, if an allrounder wasn't delivering a typical bowling output then you would at least reduce their bowling workload over time and make sure they were in as a batsman.

- The closest thing to a 'true' balanced allrounder that exists is probably Shakib Al Hasan, who is 1.03 times a good as a typical batsman and 0.966 times as good as a typical bowler

- If you had the same batting average as bowling average, and be completely 'balanced', you'd have a batting and bowling average of 35.18 (0.91 times as good as the typical batsman and bowler). That seems a little high at first, but this is 21st century averages we're talking about, where 35 is pretty much the minimum you can get away with as a batsman now. 35 with bat and ball is pretty much Shane Watson's record, curiously.
 

Sachin114

Cricket Spectator
This should be done separately for each era so there are less discrepancies in the averages.

From 1 Jan 1990 to 1 Jan 2010:
Code:
Batting	Bowling	Index
74.18	66	2
70.471	62.7	1.9
66.762	59.4	1.8
63.053	56.1	1.7
59.344	52.8	1.6
55.635	49.5	1.5
51.926	46.2	1.4
48.217	42.9	1.3
44.508	39.6	1.2
40.799	36.3	1.1
37.09	33	1
33.381	29.7	0.9
29.672	26.4	0.8
25.963	23.1	0.7
22.254	19.8	0.6
18.545	16.5	0.5
From 1 Jan 1970 to 1 Jan 1990:
Code:
Batting	Bowling	Index
72.48	64.02	2
68.856	60.819	1.9
65.232	57.618	1.8
61.608	54.417	1.7
57.984	51.216	1.6
54.36	48.015	1.5
50.736	44.814	1.4
47.112	41.613	1.3
43.488	38.412	1.2
39.864	35.211	1.1
36.24	32.01	1
32.616	28.809	0.9
28.992	25.608	0.8
25.368	22.407	0.7
21.744	19.206	0.6
18.12	16.005	0.5
From 1 Jan 1950 to 1 Jan 1970:
Code:
Batting	Bowling	Index
69.08	60.88	2
65.626	57.836	1.9
62.172	54.792	1.8
58.718	51.748	1.7
55.264	48.704	1.6
51.81	45.66	1.5
48.356	42.616	1.4
44.902	39.572	1.3
41.448	36.528	1.2
37.994	33.484	1.1
34.54	30.44	1
31.086	27.396	0.9
27.632	24.352	0.8
24.178	21.308	0.7
20.724	18.264	0.6
17.27	15.22	0.5
From 1 Jan 1930 to 1 Jan 1950:
Code:
Batting	Bowling	Index
77.26	66.44	2
73.397	63.118	1.9
69.534	59.796	1.8
65.671	56.474	1.7
61.808	53.152	1.6
57.945	49.83	1.5
54.082	46.508	1.4
50.219	43.186	1.3
46.356	39.864	1.2
42.493	36.542	1.1
38.63	33.22	1
34.767	29.898	0.9
30.904	26.576	0.8
27.041	23.254	0.7
23.178	19.932	0.6
19.315	16.61	0.5
 
Last edited:

indiaholic

International Captain
Intuitively, I would take a approach like: if the percentage of batsmen who average around 45 is approximately equal to the % of bowlers who average around 26 then those batting and bowling averages will be equivalent.
 

Sachin114

Cricket Spectator
Intuitively, I would take a approach like: if the percentage of batsmen who average around 45 is approximately equal to the % of bowlers who average around 26 then those batting and bowling averages will be equivalent.
There are 93 batsman averaging at least 50 out of 2012 batsman who have batted within the top 7, which is 4.62%.

Batting records | Test matches | Cricinfo Statsguru | ESPN Cricinfo

Batting records | Test matches | Cricinfo Statsguru | ESPN Cricinfo

There have been 19 bowlers with averages below 21 out of 366 bowlers with at least 50 wickets, which is 5.19%.

Bowling records | Test matches | Cricinfo Statsguru | ESPN Cricinfo

Bowling records | Test matches | Cricinfo Statsguru | ESPN Cricinfo

So roughly a batting average of 50 is equivalent to a bowling average of 21, in terms of how frequently it is achieved.
 
Last edited:

indiaholic

International Captain
For the period of 1990 to 2010, the equivalent bowling number seems to increase to 23. 7 out of 131 bowlers and 36 of 627 batsmen.
 

smash84

The Tiger King
I am guessing many of the sub 21 bowlers are from the latw 19th or early 20th century.

Otherwise even McG has an avg of 21.64.
 

G.I.Joe

International Coach
I attempted a sliding scale of equivalence a year or so back under the premise that the stats for the top x bowlers are equivalent to the stats for the top x batsmen. To an extent it removes the need for adjusting for era averages.

This is what I got. It's extremely fuzzy for the top 6 or so players because of the small sample size and outliers, but seems pretty acceptable x=7 onwards.

Batting average of 50 is equivalent to a bowling average of 24 in this system. Batting average of 40 is equivalent to a bowling average of 30.5.


 
Last edited:

weldone

Hall of Fame Member
I did this long ago. Remember everyone has to bat, but not everyone has to bowl. So, we can't compare an average batsman with an average bowler. Although a batting average of 10 won't get anyone selected as a batsman, but you'll see many players with batting average of 10 (because they are bowlers). You won't see similar thing with bowling, because someone with bowling average >100 won't generally bowl. For this reason, approximately an average top 7 batsman (who gets selected at least partly because of his batting) maybe compared with an average bowler (who gets selected at least partly as a bowler). An average top 7 batsman averages 36.10, and an average bowler averages 31.92. These 2 should be roughly comparable (ignoring era effect blah blah). And after that I apply simply a multiplication formula (e.g. someone who averages 36.1*1.5 with the bat is roughly equal to someone who averages 31.92/1.5 with the ball, and so on)

Batting Average ~ Bowling Average
4 ~ 319
7 ~ 160
11 ~ 106
14 ~ 80
18 ~ 64
22 ~ 53
25 ~ 46
29 ~ 40
32 ~ 35
36 ~ 32
40 ~ 29
43 ~ 27
47 ~ 25
51 ~ 23
54 ~ 21
58 ~ 20
61 ~ 19
65 ~ 18
69 ~ 17
72 ~ 16
76 ~ 15
79 ~ 15
83 ~ 14
87 ~ 13
90 ~ 13
94 ~ 12
97 ~ 12
101 ~ 11

Edit: Oops didn't notice Howe is saying almost the same thing but with different numbers lol.
 
Last edited:

smash84

The Tiger King
I attempted a sliding scale of equivalence a year or so back under the premise that the stats for the top x bowlers are equivalent to the stats for the top x batsmen. To an extent it removes the need for adjusting for era averages.

This is what I got. It's extremely fuzzy for the top 6 or so players because of the small sample size and outliers, but seems pretty acceptable x=7 onwards.

Batting average of 50 is equivalent to a bowling average of 24 in this system. Batting average of 40 is equivalent to a bowling average of 30.5.


This looks pretty good actually

:thumbsup:
 

Slifer

International Captain
Imo averaging at or less than 25 for a pace man, is roughly equivalent to averaging at or above 50 for a batsman. I'd give spinners a little more Lee way.
 

Dan

Hall of Fame Member
I think that those two numbers are probably close to equivalence overall, weldone. But in this current era of corporate 5-day roads, there is no ****ing way 47 = 25.
 

indiaholic

International Captain
I think that those two numbers are probably close to equivalence overall, weldone. But in this current era of corporate 5-day roads, there is no ****ing way 47 = 25.
From 1st Jan 2001 to 31st Dec 2010, of the 105 batsmen who scored more than a thousand runs while batting in the top 7, 25 averaged above 50. To attain an equivalent percentage for bowlers, if you set the limit as at least 50 wickets, then the average will be 29 (18 out of 79 bowlers).

From 1st Jan 1991 to 31st Dec 2000, of the 83 batsmen who scored more than a 1000 runs while batting in the top 7, 5 averaged above 50. To attain an equivalent percentage for bowlers, with the minimum at 50 wickets, the average will be 21.5 (4 out of 62)

Kinda striking IMO.
 
Last edited:

Garson007

State Vice-Captain
I don't like the linear scale of the equation. A batsman averaging 40 isn't twice as good as one averaging 20. He is many more times as good. Same applies for a bowler.

It would be interesting to see if the mean and median correlate - and if it does, if it would fit a normal distribution.
 

Spark

Global Moderator
I don't like the linear scale of the equation. A batsman averaging 40 isn't twice as good as one averaging 20. He is many more times as good. Same applies for a bowler.

It would be interesting to see if the mean and median correlate - and if it does, if it would fit a normal distribution.
This is why I think harsh.ag's approach outlined in the post above yours looks more promising to me.
 

Top