[jillybean]

2/1. It's such a gamble playing 1M 1nt* 2x 3M to show a limit raise. You've seen my openings.

I'm looking at a different approach where there is no constructive raise, 1M:2M is 3 card support

1M 1nt no 3 card support
1M 2C* limit raise or generic GF. (1M:2D = 5 cards)

1♠:2♣*
2♠ (trash): any movement by responder is gf

1♠:2♣
2♦*: (would accept a limit raise) 2♠ responder has gf, 3♠ limit with an outside ace, 4♠ limit no outside ace

1♠:2♣*
2♦*:2♠ strongest sequence
next step is show or ask shortness

1M:2nt (15+)

1M:2♣ can contain 12-14 4 card support and see what opener has

[mike777]

Playing constructive raises, three card limit raises seem to be almost non-existent in practice. Now add in active opponents. Roughly 55-60% of all auctions are contested.

Seems a lot of work to solve an infrequent problem.

[akwoo]

This is a good idea, but make sure you discuss what continuations mean if 4th seat comes in with a bid or a double.

Partner (first-time but we have a common partner) wanted to play something like this my last live game back in January (though he prefers 2♦ as turning down the limit raise), but with 10 minutes left before game time we didn't have time to discuss what to do over 4th seat interference and I didn't want to play it without such discussion.

[DavidKok]

I think this is a bad idea.

There's a lot that can be said about this, and this will be a long post. I've played similar methods for a few years, and I did not realise just how much it was costing me until I dropped it and put the limit raise somewhere else. I'll try to restrict it to three main themes: the lack of free lunch, what I consider to be the flaws in this particular set of continuations, and some available alternatives. The first of the three is most important because it speaks to the fragility of the initial 2♣ response, striking the core of the method.

1. No free lunch.
I think putting non-game forcing hands in a 2/1 auction is a bad idea for three primary reasons, and a bunch of secondary ones. The big ones are:

• With a game force no longer established, opener needs to revert to all the bean counting of the non-GF 2/1 methods. It is no longer comfortable to show shape. You're shifting this into 2♦ so that responder can show the hand, but when can opener show the hand? A big advantage of 2/1 GF is that both sides can start describing their hand, in particular their shape, cheaply. You've limited opener in their ability to do this so that responder can insert a limit raise, which is a poor bet just from the frequencies alone. I'd love to see some example auctions, especially on 1♥-2♣ (or do you intend to play different 2/1 methods after 1♥ and 1♠?) where responder has a non-minimum with extra shape, or natural diamonds - presumably you bid 2♦ first and then start describing shape at the 3-level if responder has a non-fitting game force (the most common hand type in 2♣)?
  
• A 2/1 GF turns on some gadgets in competitive auctions, in particular forcing pass. I know you 'never' have competitive auctions after a 2/1 GF, but now that the limit raise is included (where responder is weaker so there's more strength outstanding for the opponents to hold and our side has a major suit fit so the opponents have more outside cards on average) this scenario is more common. I love to overcall in fourth seat after opponents make one of these two-way 2/1 bids, for some reason they tend to completely forget you're allowed to do that. For example, what are your agreements after 1♥-(P)-2♣-(2♠); ?. Or, being cheeky about it, on 1♠-(P)-2♣-(2♦); ?, is 2♠ your weak bid, or is pass, and what about double? And what if sixth seat raises? Responder has neither shown the fit nor shown extra strength, and must hold one of the two (or both).
  
• Opener's hand evaluation is going to be more difficult if you have to assume that 2♣ is a limit raise initially. I struggled with this a lot in a style where I opened somewhat aggressively, especially with shape. When opener has one of these 'low hcp high playing strength in the major suit' hands you really shouldn't rebid 2M as you'd like to accept an invitation, but this means that the 2♦ 'positive' response also contains some 11-hcp hands and the likes. It is difficult for the partnership to judge game versus slam decisions if 2♦ says little about strength and nothing about shape, but only about some combined inference.

2. The proposed overloaded 2♦.
Secondly, I think these particular continuations are not great. On your 1♠-2♣; 2♦ I would not split limit raises into 3♠ and 4♠ based on the presence of an ace. There's another problem here which I think you've already identified: if opener wants to be in game but has no slam interest opposite a limit raise we want to bid there quickly without leaking information, whereas if opener has slam interest even opposite a minimum limit raise we want to save space and communicate more. For this reason you shouldn't jump to 4♠ with the limit raise. I think bidding 3♠ always is probably better than this split, though it won't matter much.
As mentioned I don't like the 2♦ catchall bid. I understand that your goal is to get out at 2M with a limit raise opposite a minimum opening, and on 1♥-2♣ in particular the 2♦ bid is the only way to sensibly do this, but I'd still permit opener to start describing immediately with hands that both have shape and would bid game opposite a limit raise. However, there are still hands that slip through the cracks because of the heterogeneity of the 2♣ bid, such as ♠KQxxx, ♥x, ♦AKTxx, ♣xx. This is an easy game bid opposite a limit raise, but on 1♠-2♣; ? you might not want to bid a nebulous 2♦ and also not bid a natural 3♦, for example (despite both having shape and accepting a game try). I would still split out more options from the 2♦, this catch-all 'playing strength showing opposite a limit raise' bid hampers communication.

3. Some alternatives.
Lastly some alternatives. I play Maas 2NT, which commits us to 3M with a limit raise opposite a minimum opening. In my experience having the Maas 2NT is a huge winner, much better than putting the limit raise in a semiforcing NT or artificial 2/1, and much better than Jacoby 2NT. Compared to your post title it primarily shifts the inferences around though, it does not solve any 3M-1 problems. At the same time though, these 3M-1 auctions are rare and sometimes even gain because it was a good sacrifice over their 2♠ or 3-level contract, and I've even seen it keep opponents from finding their own game.
One alternative that I've primarily seen in relay context is to use 2♣ as a catch-all game force (some relay systems even use 1NT or 1♠ instead), freeing up other 2-level responses for artificial uses. A common one is transfers, so:
1♥-?

• 1♠: standard
  
• 1NT: standard (e.g. forcing or semiforcing)
  
• 2♣: Game forcing relay
  
• 2♦: Limit raise in hearts
  
• 2♥: Weaker hearts raise

The structure over 1♠ is similar though you need to decide whether you want '2♦ = hearts, 2♥ = limit raise in spades' or the other way around. Personally I think these systems also have significant flaws, and that the 1M-2♣ game forcing relay is one of the weaker relay sequences out there (which is saying something - I think relay is very overrated in general). However, I've seen it argued that taking a loss on the 2♣ auctions is acceptable because the 2-level limit raise bid allows us to push the 1M openings even lower, which benefits on many competitive sequences and constructive sequences where the relay is not in play.
Thirdly you can insert constructive raises. Personally I'm not a fan of these, but they significantly reduce the frequency of responder holding a true limit raise while also letting you stop in 2M regularly.


In general the limit raise is a hand type that slips between the cracks of common systems - note that the same problem exists outside 2/1 GF systems. If you open light, in the modern style, and a limit raise opposite makes 3M a bad contract this has the potential to be costly. However, it is somewhat infrequent, need not be costly when it comes up, and finding room in your system below 2M is really rough. I think most of these cures listed above are worse than the disease, and that's why I play Maas 2NT (primarily for its other upsides) and accept the occasional 3M-1.