BBO Discussion Forums: Acol - BBO Discussion Forums

 pescetom, on 2023-April-12, 11:50, said:

Probably depends on your follow ups to 1♣. If we sometimes miss a spade fit because 1♠ is fourth suit, or if we play Walsh, I can imagine that 5533 is marginally better. Anyway it surprises me that the difference is big enough to reach statistical significance. Unless he has a sample size of a billion.

The difference between flavours of 5cM isn't very big, and he explains that anything under approximately 0.5 IMPs per 24 board match should be dismissed. In the relevant comparison (of 300 24-board matches, so 7,200 boards) the systems scored:

5533 +0.67; 4444 HSCD +0.25; 5542 +0.18; 4444 HCSD +0.00; 4444 CDHS -0.43; 5551 -0.68.

I would not be surprised if the continuations of 5542 do not resemble modern expert systems, but the book doesn't go into further detail. My personal interpretation is that all these differences are small.
Just for comparison (and because I can't resist), he compares strong (15-17), variable (weak only when NV) and weak (12-14) NT on the same scale:

5533 Strong NT +1.67, 4444 Strong NT +0.97, 5533 Variable +0.44, 5533 Weak -0.22, 4444 Variable -1.08, 4444 Weak -1.78.
My interpretation is: the more frequently you play weak NT, the more IMPs you're handing over.

 DavidKok, on 2023-April-12, 15:27, said:

Thanks.

I was uneasy with Acol 4cM and weak NT, but then I had no idea what I was doing at the time.
I felt more confident with Italian 4cM and strong NT, even better with 5cM and strong NT: never tried 5cM and weak NT yet, have an open mind on it.

I briefly played 5533 in two flavours and then switched to 5542 with no regrets.
I agree that the differences are small (especially if 5533 can only be 3 card diamonds with 4=4=3=2) but my tentative conclusion is that they are consistently favourable to 5542 and I find it hard to imagine the contrary.
The book will be part of my reading this summer in any case.

I haven't read the book, so maybe I shouldn't criticise it

But it seems to me to be a bit meaningless to compare all those opening schemes if you don't specify in details how the systems are worked out. Maybe the systems are just tuned to strong notrump, for example by a 1NT response to 1x is defined as 5-10 points. Or stuff like inverted minors, shape requirements for 1NT rebids, check-back method, requirements for negative doubles etc etc. Not to mention opponent's overcall style and NT defence. The fact that the analysis is double dummy also means that some of the advantages of aggressive systems are lost.

There is much more bidding theory and much more software developed for strong NT than for weak NT. Maybe the results just reflect that?

Or maybe opps are very timid in the sims. If opps will rarely interfere over you 1m openings it is obviously better to open 1m than 1NT

 P_Marlowe, on 2018-September-24, 04:25, said:

I don't think this is right. See Standard English Acol, System File 2020. And Beginning Bridge Book One

 rienzi, on 2023-April-13, 08:08, said:

It looks a pretty good defining feature to me.

Even if we accept that the EBU owns Acol (it doesn't) the Standard English System File that you quote has: "A change-of-suit response at the one level promises at least six points and introducing a new suit at the two level at least 10 points. There is no upper limit. Opener must bid again (unless, exceptionally, responder has already passed).".

The requirement for opener to bid again makes the bid forcing and the 10-point requirement is certainly not enough to force to game, so the response is forcing for one round.

I would say that it is one defining feature.
If it was "the" defining feature, then about half the natural systems in the world including Standard American would be Acol.

 pescetom, on 2023-April-13, 10:35, said:

Actually, Dutch Acol (strong NT, often played as 5-card majors) is probably identical to what SA would be if played with Acolish forcing character of the 2/1 responses.

On the other hand, if you play 2/1 as auto-forcing (as in SAYC and SEF) I don't think many people would refer to the system as "Acol".

 helene_t, on 2023-April-13, 11:17, said:

I don't think many people would refer to Italian "Lungo Corto" as "Acol" either, although it shares the key characteristics of 2/1 as a simple force and 4 card major openings, in this case combined with strong NT. In many ways it was a bug-fix of Acol, but nevertheless died abruptly once the Italian variant of 2/1 5cM emerged.

Posters might look at "The Science", a 4 card Major 14-16 2/1 system. It is fun.For many years, beginning with Baronised Acol. a 2/1 was forcing to 2NT. Very few play that responder does not need to bid again.

 the hog, on 2023-April-13, 17:09, said:

Almost everybody in the UK plays 1♥-2♣-2♥ is NF but anything other than opener repeating his suit requires responder to bid again. 1♥-2♣-2♦-2♥ is also NF.

 Cyberyeti, on 2023-April-14, 01:36, said:

Seriously, do you always have to make a comment and be a contrarian? Actually you are quite incorrect. Perhaps beginners and weak club players play this as nf, but strong Acolites do not. Having said this, there are not that many strong pairs in the UK who still play Acol.

 the hog, on 2023-April-14, 02:12, said:

I would agree with Cyberyeti both are NF. Opener can have a 10 count and responder a 9 count.

 TMorris, on 2023-April-14, 02:30, said:

"I would agree with Cyberyeti both are NF." So you are an expert?

 Cyberyeti, on 2023-April-14, 01:36, said:

I think
1♥-2♣
3♣
is also nonforcing for most. Which means that if you play that a 2434 opens 1♥ (as in "Standard English") you probably don't want to raise clubs (also because that would suggest a fifth heart), you rebid 2NT and the club fit might never surface.

 the hog, on 2023-April-14, 02:38, said:

No, nor have I suggested I was. What of it. It is common for these auctions to be NF in Acol. Some may play them otherwise. I play regularly with experts and am confident they would consider these auctions NF.

 the hog, on 2023-April-14, 02:12, said:

I am not sure which strong Acol players you are referring to. Maybe the two Acol pairs from the Welsh team that won the Camrose in 2011, I think they were the last Acol pairs that won something major although there might have been others which I didn't notice. They both state "2/1 9+" on their system card without any clarification about forcing character. If 2/1 is 9+ it seems unlikely to me that it promises a rebid, but I suppose it is possible.

We have one strongish Acol pair (has won an inter-county and various congress events) at the local club and they play as Cyberyeti explains. In NZ we had an Acol pair at the local club who were on the team that won the NZ county teams in 2020, they also play that style.

If 2/1 promises a rebid I don't think anyone would call the system Acol here in the UK or NZ. Maybe it's different in OZ, I wouldn't know.

Then again, I don't think I have ever come across a natural weak4 system with 2/1 autoforcing, it seems unplayable to me as you will be at the 2-level with a combined 20 points and a misfit so you need to stop ASAP.

 pescetom, on 2023-April-13, 14:58, said:

It may just be a question of whether the system has historically evolved from Acol or not. Dutch Acol branched off from English Acol at a time when English Acol was mostly played with a variable NT, so it was similar to how the English would play when vulnerable. So they still call it Acol, or at least until recently (I think most teachers stopped calling it Acol when they switched to 5cM, on the other hand some 15-20 years ago most club players would learn 4cM but play 5cM, and both would be called Acol).

On the other hand, if you are in North America and you play Goren with limit bids (and maybe 5cM), there is no particular reason to call it Acol even though it may be identical to Dutch Acol.

 the hog, on 2023-April-14, 02:12, said:

You have no clue about what is played in the UK, when I played in the last 8 of the gold cup (UK's prime KO, 48-64 board matches) all 3 pairs in our team played this way.

 helene_t, on 2023-April-14, 02:43, said:

You don't necessarily lose the club suit. One method has the sequence 1♥-2♣; 3NT as showing this hand-type with club support. Alternatively, responder can rebid a forcing 3♣ over 2NT (if not playing Check-back), if responder is interested in playing in clubs opposite a fitting club suit.