New ways to rate cricketing talent.

[RazabQ]

Anyone use CI widgets? perhaps we can find out if they use SOAP or something else to get the data

[al Furqaan]

Originally Posted by Neel Here
furqaan bhai link to WEIR please ?

http://www.banglacricket.com/alochon...highlight=weir

however, here are the formulas:

Batting Formula (ODIs):

(total runs) * (strike rate)
______________________
(innings) * 100


Batting Forumala (Tests):

is the average of ODI formula and:

(total runs)
_________________
(innings) * 100


Bowling Formula (ODIs):

(total wickets) * (60,000)
_________________________________________
(innings) * (econ) * (average) * (strike rate)



Bowling Formula (Tests):

(total wickets) * (20,000)
_________________________________________
(innings * 2) * (average) * (strike rate)

[RazabQ]

Wait al, I think Neel thinks you are older than me! You get the bhai, I get the 1st name! LOLO

[Neel Here]

oops ! sorry bhai !

[Neel Here]

Limited Overs Individual Batting Performance Evaluator (LOvI BatPor)
(to bangla speakers, figure out what that means !)

(applicable for 50-50 and T-20)

Acronyms Used : --

R -- Runs Scored by a batsman in an innings.
SR -- Strike Rate of the batsman's innings.
InSR -- Innings SR of the batsman's own team. *
MSR -- Match SR, of both teams combined. *
InAvg -- innings average of batsman's own team, calculated by dividing final score of his team minus extras /no of wkts fallen.
MAvg -- match average, average individual score of both teams combined.

* In case a team is all out and fails to complete it's full quota of overs, SR will be calculated assuming that full quota of overs has been played. For teams batting second completing the chase in less than full quota however, the exact no of overs batted will be considered.

------------------------------------------------
In the LOvI BatPor system, a batsman is awarded points on three criteria

A ) Hit and Run Score (HRS)

In a limited overs match, both runs scored AND SR are important but the level of importance changes according to the total number of overs.
Strike Rate is a more important metric in say, a T-20 match than a 50-50 one.
This criteria will try to incorporate both these factors in a quantitative way using weights.

Calculation Formula


Here, is the weight used in order to assign importance to runs scored and the pace at which they are scored.
In my view, a proper value would be :
o for 50-50
and
o for T-20

the formula then becomes :
50-50


and
T-20


Comments :-- Mostly Self explanatory, the formula would perhaps become clearer with an example, I've calculated some but it takes time to post all that !
For the moment, members can tweak the weights to see if they see fit.

----------------------------------------------

B ) Quality of Opposition Team Score (QOTS)

Objective of this score will be to reward performance against good teams and thereby even out scores against weaker teams.
Determined on Two Counts, difference in strike rate(which can be seen as a measure of dominance over bowling) and difference in runs scored and average.

i) QOTS Strike Rate Difference


ii) QOTS Average Score Difference


Comments :-- These two parameters compare the performance of the batsman wrt other members of his team, in doing so it is assumed that against a good opposition there would be in general lesser no of good performances than against weaker teams.

Therefore, this parameter will
o reward him for performing better than his teammates, IOW performing against
quality opposition.
o neither reward nor punish him for performance at same level as his teammates
o subtract points for performance worse than that of his teammates.

-------------------------------------

C ) Ground Condition Score (GCS)

How to account for a seaming greentop where a batsman scores a gritty 75 while his team is all out for 200 and the opposition too bundles out for a similar low score ?
it is certainly not the same as a 75 where his team scores 200 in reply to oppositions 350 !
and it is much more valuable than a 125 on a flat wicket where his team scores 300 + !

the way to do it would be to compare his performance with scores from both teams
A very good team playing against a weak team can still skew up the performance of one team but using data from both teams, we can figure out if the weather and pitch had some role to play after all or not !!

Again, we calculate this score on two counts but this time, we use the match averages and SR :
i) GCS Strike Rate Difference



ii) GCS Average Score Difference



-------------------------------------------

That's it for the day ! feel free to pick holes and/or suggest improvements !

Coming Next : Score of some batsmen from the famous Aus vs RSA run chase, followed by Bowling Analysis System.

Edit : Not Out Bonus :-- A batsman staying not out is always advantageous to his team, the longer a batsman has stayed out and more the runs he has scored, the better it is for his team.
a NO batsman is awarded with 0.2 points for every run scored.

[RazabQ]

Damn Neel, eita porte shomoy laagbe - I'll come back and check later in the evening.

[al Furqaan]

the formula must also be tested so that its objectivity correlates reasonably with our subjective accepted conception of various players relative ability.

* In case a team is all out and fails to complete it's full quota of overs, SR will be calculated assuming that full quota of overs has been played. For teams batting second completing the chase in less than full quota however, the exact no of overs batted will be considered.

don't agree with this part.

since its the SR, it should be runs/balls ratio, period. no need to extend it to the full quota of overs. thats a NRR calculation, which has its flaws. basically we are rewarding our batsman for the good work of the bowlers (or misdeeds of the oppossing batsman). the batsman's stats should be simply their own handiwork. if say a team scores 200 in 45 overs all out, their SR will be calculated at 66.67 rather than their real striking rate of 74.07.

[Neel Here]

the formula must also be tested so that its objectivity correlates reasonably with our subjective accepted conception of various players relative ability.

yes, it is still being created, fully open to tweaks and suggestions.

that is one reason why I was calculating examples, to calibrate the system. no doubt there's still work to be done but the figures I got till now look promising.

[al Furqaan]

Neel, do have a single formula that encompasses pitch conditions and strength of opposition??

[Ajfar]

neel bhai..after seeing this my calc homework looks like a peice of cake

[SGW]

Nice work Neel - I think you are onto something good there...

[al Furqaan]

neel bhai,

your formulas are for a single innings or match...is there a way to get a cummulative rating for a series/season/career?

also, try and finish ur system...we can run it up against my WEIR and see if the results coincide. the WEIR is a much simpler method tho...

[al Furqaan]

btw, hearty BC welcome to Coach Williams!!!

i hope you can contribute to BC for years to come!

[MarufH]

Originally Posted by SGW
I'm interested to see what else you guys might think about how we can measure players OBJECTIVELY.....it's hard to measure the intangibles......because there are so many, and still rely on "opinion" as opposed to "fact"...

Anyway, off subject for anyone who might be interested.....Enam will not play for Maharashtra in the upcoming series....his first priority is to Bangladesh cricket of course.....

As for Maharashtra, things here are going well....we have maintained our position in the Elite League of the Ranji Trophy and we currently have by far the youngest squad in the competition (average age is 21.5)....we are in the process of building a strong side for the future and part of that building process is giving our players exposure to international tours such as the upcoming series in Bangladesh. We will also be touring Australia in June/July after visiting Bangladesh...

So we're really looking forward to coming to B'desh and taking part in what we hope will be a very competitive series of matches....personally I'm also looking forward to catching up with many of the players again and seeing some old friends in Bangladesh once more......

Should be good fun

Best Regards
Shaun

Thanks for clearing that out! Also, I am glad things are working out for you in India...

Coach... see if u can recruit Mithun Ali or Boro Shuvo for Moharastra..... these kids are results of your hard work!!! They need proper exposure!

[MarufH]

Mods/Admins... can you guys change his street cricketer status into Former BD coach?

[bujhee kom]

Originally Posted by MarufH
Mods/Admins... can you guys change his street cricketer status into Former BD coach?

I second that dear Mod Admin bhais please!
Thank you dear Maruf bhai for pointing it out!
Honoring coach SGW is honoring the BD tigers

[Neel Here]

quick question ! how do I put in a scorecard ?

like in this post ??
http://www.banglacricket.com/alochon...&postcount=179

[Neel Here]

Ricky Ponting Vs Herschelle Gibbs.


SA vs Aus (ODI no. 2349)
http://content.cricinfo.com/ci/engine/match/238200.html

Played at New Wanderers Stadium, Johannesburg, on 12 March 2006 (50-over match)

Result South Africa won by 1 wicket (with 1 ball remaining)
__________________________________________________ _____________


INSERT SCORECARD HERE. (someone please tell me how to do this.)

__________________________________________________ _____________

Some Calculated Figures :--
Team...........Strike Rates........Batting Avg.
Aus............144.67 (In.SR)......103.75 (In.Avg)
SA ............146.5 (In.SR).......46.44 (In.Avg)
Both...........145.58 (M.S.R).......64.08 (M.Avg)

==================================================
Now let's compare the two very similar performances from both teams,

Ponting's 164(105) @ 156.19
and
Gibbs 175(111) @ 157.65

Player.....Runs.....S.Rate....H.R.Score....QOTS(SR)....QOTS(Avg)....GCS(SR)....GCS(Avg)....Total Score

Ponting....164.....156.19......187..........18.89........ ..60.25.........17.4..........99.92.........383.46
H.Gibbs....175.....157.65.....200.22.......19.50......... 128.56.........21.1........110.92.........480.3
================================================== =====================

In most rating systems, gibbs and ponting would have got very similar scores, but here gibbs clearly has a big advantage.
In my opinion that reflects positively on the systems ability to take into account the
difference in the bowling sides and match situation.
But I'll not analyze the results now.

I'm asking BCites to first analyze the results (hint: focus on the individual components ) and say what they think !
It's better that we can have some unbiased opinion before looking at my analysis, which would surely have some personal bias.

[Neel Here]

T_E bhai's message.
posted with permission.

Originally Posted by Tigers_eye
Hello Neel,
I went through your performance evaluation system, and must say excellent thought process was put in.

1) To have any type of rating one has to have a level playing field. The scores Gibbs and Ricky "I am the forth Umpire" Ponting have 480 and 380 are open ended.

2) Ground condition, opponent strength has equal values as runs and SR. Those parameters are important but not the most important parameter.

3) Bating position has not been considered. (Even though both were top order batsman) In future, we would compare between Bhua and Stud Shakib's innings.

So my suggestion is make the rating a weighted score:

H.R.Score(75%)..QOTS(SR)(5%)..QOTS(Avg)(5%)..GCS(SR)(5%)..GCS(Avg)(5%)..BP & NO(Batting postion & not out)(5%)

and my reply

T_E bhai, don't you think the qots and gcs deserve more weightage ?
as a matter of fact I am NOT YET done on this, I'm tempted to scale the whole QOTS/GCS thing to a par score. will post later on that.

among other things, can I post this PM on the thread ? it has some points that are worth discussing.
regards.

.....................................
furqaan bhai, I'll reply in detail later, tomorrow hopefully. for the moment let me just say that you can actually combine the opposition team and ground condition in a single formula, it won't change anything but that make understanding the points difficult.
I would look forward to your comments on the gibbs vs ponting example.

to all who have commented, thanks for the kind words !
even more than praise, please keep the criticism coming !

[Zeeshan]

neel bhai,

genius thread... was wondering if you could post ratings for a gradient from crappy players to Sachin tendulkar and give us a realistic glimpse of their talents.

[cricman]

Using the Coaches System I did All the Batsmen up to date (No Fielding & Bowling). I didn't Rewards players for Not Outs. The Results are per Innings

Ashraful: 31.102
Shakib: 33.72
Mushifiq: 28.35
Tamim: 31.21

If you want it per Test Match just * 2.

Shakib would be pretty close to 150ish I would assume if you included his fielding and bowling.

125 Would be Test Class from my prespective, Ash and Tamim are around 70-80's and Mushifiq close 100 only due to the opportunites he gets for dismissals.

[SGW]

Like your work Neel - obviously you've put a lot of work into that (!)
Can I ask you how practical and how easy is it for you to punch in the numbers - e.g How long would it take you to rate every player in one match or one series for example (?)
Keep it up though - it's really interesting stuff.....
Regards
Shaun

[Neel Here]

thanks for the response shaun !

it takes about a minute for calculation of the scores for each player with the calculator.
you'll notice that a few stats are common for all players (innings SR, innings avg, match SR etc) and the input data for the individual players (SR and runs scored)can be simply read off the score cards.
it will also be fairly easy to write a code for this, it would be a very simple one, 20-30 lines at most. that can bring the calculation time for a whole match to about a minute !

as a matter of fact it takes faaaaar longer to type it down in a readable format for BCites than to calculate it !

[al Furqaan]

also, Neel if its possible to scale your system to 100 (actually 99) please?

this is has the benefit that its the number system we use (if we were Mayans i'd suggest it to be scaled to 60). by this i mean, once you have every player done...pick the highest rated player (Tendulkar, KP, Kallis) whoever it might be, and set his career score at 99. basically this could let us calculate percentages very easily. of course the guy at 99 might start jumping over 100 based on performance, but if you pick one of the guys i mentioned above, it won't be too great a deviation.

thanks and good work.

i will try to run the SL-BD U19 series into my WEIR and lets see what happens when we compare your complex system vs my simpler one.

cheers and carry on the hard work!

[Zeeshan]

Originally Posted by al Furqaan
also, Neel if its possible to scale your system to 100 (actually 99) please?

this is has the benefit that its the number system we use (if we were Mayans i'd suggest it to be scaled to 60)....

Mayans used vigesimal or base 20 system. you are confusing them with the Babylonians.

And I respectfully disagree making the scaling out of 100 because then it looks like master-moshai's nomboring system.... Rather, in the interest of preserving individuality I prefer the credit score or Chess rating type of system with no limits.

Just my two pence...