West Indies: Another team that doesn't get the 750 concept

[tendulkar.com]

now facing the possibility of a defeat.

Note: In many cases, you may actually draw or win from these situations, but you are giving your opposition a higher chance (say from 1% win to 10% win), which is bad strategy.

We are talking about strategies which gives you statistically better chances to win.

[Andrew Dunford]

"tendulkar.com" <qro...@gmail.com> wrote in message
news:3c78fc4a-0d72-4887...@googlegroups.com...

Ok, I'll bite. Where are these statistics?

Andrew

[tendulkar.com]

I've provided proof by reasoning on various times(which is more solid than statistics). You should just search for my journal paper on this :)

if x + 2 = 4, then you don't need statistics to prove x=2.

You take the statistics route when you can't reason about it.

[patt...@gmail.com]

Except you've posited x + y = 4 and assumed y = 2.

[alvey]

On Mon, 19 Nov 2012 08:22:52 -0800 (PST), tendulkar.com wrote:

> On Sunday, November 18, 2012 4:07:00 PM UTC-5, Andrew Dunford wrote:
>> "tendulkar.com" <qro...@gmail.com> wrote in message
>>
>> news:3c78fc4a-0d72-4887...@googlegroups.com...
>>
>>> now facing the possibility of a defeat.
>>
>>>
>>
>>> Note: In many cases, you may actually draw or win from these situations,
>>
>>> but you are giving your opposition a higher chance (say from 1% win to 10%
>>
>>> win), which is bad strategy.
>>
>>>
>>
>>> We are talking about strategies which gives you statistically better
>>
>>> chances to win.
>>
>>
>>
>> Ok, I'll bite. Where are these statistics?
>>
>>
>>
>> Andrew
>
> I've provided proof by reasoning on various times(which is more solid than statistics).

Statistics say otherwise.

> You should just search for my journal paper on this :)
>
> if x + 2 = 4, then you don't need statistics to prove x=2.
>
> You take the statistics route when you can't reason about it.

You mean like; "(say from 1% win to 10% win)"?

And btw, have you emailed the Windies to tell them how badder their
strategy was in T1?




alvey

[Andrew Dunford]

"tendulkar.com" <qro...@gmail.com> wrote in message

news:cdba7821-0490-484d...@googlegroups.com...

I'm not going to turn this into a long-winded debate or read your journal
paper, but two things:
1. The 750-run theory is primarily about insurance against loss, not about
winning
2. Giving your opponent a "higher chance" of winning does not necessarily
decrease your own chance of winning. It simply reduces the likelihood of a
draw.

I have no doubt that following your strategy will in some cases produce draw
when victory might otherwise be achieved, and it's timely that you posted
this about the West Indies v Bangladesh match because it may well have been
just such an example.

Andrew

[tendulkar.com]

It's their problem and they should seek out me.

[tendulkar.com]

You have a point about 'draw' risk. But, Windies match actually proves my theory.

Windies eventually scored 800 runs in the match, but they were exposed to a loss. Had they declared at 750, they would have never lost the match, but had the same opportunity of taking 20 Bangladesh wickets < 750 in < 225 overs

(They would have still bowled on the 4th and 5th day, which was more helpful to bowlers)

[alvey]

On Mon, 19 Nov 2012 22:51:18 -0800 (PST), tendulkar.com wrote:


>>
>> And btw, have you emailed the Windies to tell them how badder their
>>
>> strategy was in T1?
>
> It's their problem and they should seek out me.

"reasoning" eh? How about this reasoning then?

The Windies won.
Therefore they don't have a problem.
Therefore they don't need to seek you out.




alvey

[tendulkar.com]

Just because you didn't crash doesn't mean you drove correctly.

[Unknown]

Ditty-Wah-Ditty

max.it

[alvey]

If the object of the drive was to not crash then I drove correctly.
If the object of the game was to win then the Windies played correctly.

This "reasoning" game is fun!




alvey

[tendulkar.com]

correct, only if you drive once in your life.

[Bharat Rao]

On Wednesday, November 21, 2012 12:21:12 PM UTC+5:30, tendulkar.com wrote:
>

I shouldn't, but I'll bite. Your arguments would have a lot more credibility if you responded to all critiques rather than just the ones that are trivially flawed.

What you relentlessly ignore with your 750 mantra is that is also reduces the chances of winning substantially. Mike made the very valid point that had India batted to 750 (presumably into D3), England would have had much better chances to save the game.

To compound that you said that you didn't think that the breakthroughs on D2 end were critical -- they certainly were. The only time the pitch had life was in the first session of every day -- without the breakthroughs on D2 end, I wager England would have put on a lot more in I1, making this another dull draw.

No, Dhoni got the I1 declaration down pat. Where he perhaps erred (regardless of the result) was not batting in the very comfortable conditions at the end of D3, resting his bowlers, and making England come out D4 morning with 6 sessions to bat..

> correct, only if you drive once in your life.

It is truly ironic that one who correctly criticizes those who use one-off examples, does exactly the same himself. Picking every situation that remotely fits the 750 scenario (even when it doesn't, ala India), while ignoring all those that don't fit...

Bharat

[Anant Rege]

Not sure why everybody is arguing here when statguru is sitting idle.

Score Inning Matches Won Drawn Lost
750+ 1 3 1 2 0
700+ 1 7 3 4 0
600-699 1 64 34 30 0
600-699 2 41 27 14 0

So no team scoring more than 600 after batting first or second ever
lost in the history of the game. So OP should shut up at least until
that happens.

[tendulkar.com]

Jump from the top of Empire State Building. As you pass through all the floors and at the first floor, here's how your data will look

# of floors travelled: 85
# of times hurt: 0

Conclusion at that point: Jumping from Tall Buildings is super fun.

[tendulkar.com]

Remember there is an optimal win-loss-draw ratio for a team.

Rule 1: Don't lose.
Rule 2: Win
Rule 3: Don't bat on Day 5.

(If you apply Prospect theory, http://en.wikipedia.org/wiki/Prospect_theory especially when you are dealing with winning positions, at a minimum 1 loss = 2 wins)

Ask the Australians about 2001 Calcutta. Ask McGrath/SWaugh how hurt they were. They would probably swap at least 10 meaningless home thrashings of India at home to one Calcutta victory

If you really don't understand what I'm saying,
a) set-up a Monte-Carlo simulation of these scenarios and derive the win-loss-draw ratios, (over a billion simulations).
b) Configure a game theory payout matrix (use correct payouts from behavioural science)
c) Figure out the dominant strategy.
d) ...
e) Bow to me

[alvey]

On Tue, 20 Nov 2012 22:51:12 -0800 (PST), tendulkar.com wrote:

>>
>
> correct, only if you drive once in your life.

You are the Salvador Dali of rsc.



alvey

[alvey]

On Wed, 21 Nov 2012 09:54:02 -0800 (PST), tendulkar.com wrote:


>
> Jump from the top of Empire State Building. As you pass through all the floors and at the first floor, here's how your data will look
>
> # of floors travelled: 85
> # of times hurt: 0

Actually the stats would be;
# floors "passed through" = 0

There's a floor in your statement.



alvey

[jzfredricks]

On Thursday, November 22, 2012 4:21:41 AM UTC+9, blrgr...@yahoo.com wrote:
> 750 score in a test *could* be a eye popping outlier based on the data we have - scale it down a bit and you might get more acceptance.

I guess that explains why my "1000 run rule" gets little attention :(
It took me, literally*, minutes to design and be comfortable with it. It's on par with tend.com's theory, but it has more ZEROs (so it must be better).

* I had to do some statsguru searches to back it up

[Andrew Dunford]

"tendulkar.com" <qro...@gmail.com> wrote in message

news:37f35db4-26f6-4316...@googlegroups.com...

This analogy makes no sense. A jump is an entire event - there is no
conclusion to be drawn whilst it is in progress.

Presumably you're trying to say that just because teams scoring 600 in their
first innings have never lost until now, that is no guarantee of it not
happening in the future. Of course that is true, but something quite
different from the 'jump' story.

Andrew

[Bharat Rao]

On Thursday, November 22, 2012 3:10:45 PM UTC+5:30, Andrew Dunford wrote:

> Presumably you're trying to say that just because teams scoring 600 in their
> first innings have never lost until now, that is no guarantee of it not
> happening in the future.

> Andrew

Well it could happen this Test simply because of the speed of scoring -- say Aus 600ao soon after lunch on D2. RSA get to 550 odd end of D3. Aus collapse for 180 on D4, and RSA limp home for 6 down on D5.

The fundamental problem with the 750 theory is that it is .. well ultra-simplistic. Of course, scoring 750 in I1 makes you a bit safer than scoring 600; the problem is that it also diminishes -- by several factors more -- your chances of winning the Test.

As far as I can tell looneytunes.com's 750 theory is based on the assumption that teams score about 300 / day. So go ahead and score 750, declare after lunch on D3, and you now have 2.5 days to take 20 wickets assuming you can enforce the follow-on, when the pitch is the worst for batting. And if the opposition managed 551, you weren't going to win anyway.

Seems beautifully simple when viewed from an armchair, doesn't it, but it has many flaws.

Just to name two, teams losing wickets tend to score at a much slower rate than teams not losing wickets. So if a team was in danger of losing 20 wickets in those 2.5 days, I would expect them to score 650'ish not 750'ish. The more important thing is that bowlers are not robots; they get tired. Bowling for 2.5 days straight makes you a heck of a lot less effective for the last day or so, than having an innings break. And for measure, if you do this for a 4-Test series, you will have bowler breakdown.

Far superior is to score 550-ish, or whatever you can get to an hour before close. Let your bowlers be fresh for two cracks with the new ball. Then (unlike India) do NOT enforce the followon (unless there is not enough time left). Bat, even if it is for a session or two to get to that 750 combined-total, scoring a lot faster because you have wickets. You do take a slight risk that the team may score at 4.5 rpo and overhaul the total or that you may collapse and allow a win (neither of which are possible if you scored 750 in I1), but the probability of that happening is a lot lower than the risk that your bowlers will be ineffective on D5 post-followon.

Put it this way, if strategy A increases your chances of winning by 10 percentage points over strategy B, and also increased the chances of losing by 1 percentage point, most rational humans (except Black Swan misinterpretees) would pick Strategy A.

Bharat

[arb...@gmail.com]

When I was a kid, some 25 years ago :-), my uncle - who I owe much of my interest in cricket to - would say that if a team scores 400 in either innings, they simply cannot lose. (I wonder, did the stats support his reasoning, back in 1985?)

I guess in this day and age, that's closer to 600... Certainly in the last 20 years, dozens of tests have been lost by teams that have made 400+ in either innings. Just last week Bangladesh lost at Mirpur after scoring 550 in their first innings, and England lost at Ahmedabad after making 400 in their second.

Maybe tendulkar.com is just ahead of his time. Perhaps in 2035, you'd need to score 750 to be sure not to lose.

-Samarth.

[John Hall]

In article <10025e13-895a-4df6...@googlegroups.com>,

arb...@gmail.com writes:
>When I was a kid, some 25 years ago :-), my uncle - who I owe
>much of my interest in cricket to - would say that if a team scores
>400 in either innings, they simply cannot lose. (I wonder, did the
>stats support his reasoning, back in 1985?)

<snip>

It was certainly rare, but not unknown. ISTR that in the first innings
of "Botham's match" at Headingley in 1981 Australia made 401.
--
John Hall

"Whenever people agree with me I always feel I must be wrong."
Oscar Wilde

[Andrew Dunford]

<arb...@gmail.com> wrote in message
news:10025e13-895a-4df6...@googlegroups.com...

> On Wednesday, November 21, 2012 8:29:46 AM UTC-8, Anant Rege wrote:
>> On Nov 16, 10:48 am, "tendulkar.com" <qros...@gmail.com> wrote:
>>
>> > now facing the possibility of a defeat.
>>
>> >
>>
>> > Note: In many cases, you may actually draw or win from these
>> > situations, but you are giving your opposition a higher chance (say
>> > from 1% win to 10% win), which is bad strategy.
>>
>> >
>>
>> > We are talking about strategies which gives you statistically better
>> > chances to win.
>>
>>
>>
>> Not sure why everybody is arguing here when statguru is sitting idle.
>>
>>
>>
>> Score Inning Matches Won Drawn Lost
>>
>> 750+ 1 3 1 2 0
>>
>> 700+ 1 7 3 4 0
>>
>> 600-699 1 64 34 30 0
>>
>> 600-699 2 41 27 14 0
>>
>>
>>
>> So no team scoring more than 600 after batting first or second ever
>>
>> lost in the history of the game. So OP should shut up at least until
>>
>> that happens.
>
> When I was a kid, some 25 years ago :-), my uncle - who I owe much of my
> interest in cricket to - would say that if a team scores 400 in either
> innings, they simply cannot lose. (I wonder, did the stats support his
> reasoning, back in 1985?)

It was never true as such, but perhaps truer then than now. Prior to 1985
this had occurred 38 times, and it has happened 63 times since of which 54
have been since 1 Jan 2000. I would be curious to know whether your uncle
spoke to you before or after 18 January 1985, which was when India lost to
England after scoring 412 in its second innings.

<snip>

Andrew

[Southpaw]

Likely after, since my knowledge of cricket was a bit too primitive to grasp test match strategy, in Jan 1985. Which of course didn't prevent me from attending multiple days of the Chepauk test match that you're talking about.

Evidently, when he ran his theory by me, I'd forgotten what India made in their 2nd innings, even though I was actually present at the ground for portions of that test.

-Samarth.

>
>
>
> <snip>
>
>
>
> Andrew

[Andrew Dunford]

"Southpaw" <arb...@gmail.com> wrote in message
news:c7b54cf2-ba63-4e3d...@googlegroups.com...

My experience of that adage about 400 was that it was generally applied to
the first innings of the team batting first. And still is to some extent,
often blindly, which is where Roshan has a point. Clearly the length of
time taken to score runs is one of several other factors, which is where
Australia's 482/5 in a single day is interesting.

Andrew

[alvey]

On Friday, November 23, 2012 9:35:13 AM UTC+10, Andrew Dunford wrote:
> "Southpaw" <arb...@gmail.com> wrote in message
>

>

> My experience of that adage about 400 was that it was generally applied to
>
> the first innings of the team batting first.

That "400" figure, like just about all cricket stats now, needs a weighting factor to allow for the significantly reduced playing areas of todays grounds. Imagine what The Don's average would be now. About 150 I'd hazard.



alvey