[jillybean]

As mentioned in the other thread, I have moved from over using (and abusing) splinters with most GF hands to playing 2/1 style methods where I have many ways to show gf+ hands which do not take up the entire 3 levels of bidding space. I now use a "splinter" to show a weak hand that would jump to 4M over a 1M opening, with shortness.

How do you use them?

[LBengtsson]

never weak, never strong (as will use 2Nt jacoby). so intermediate range where good trump support and shape can lead to slam

[Cyberyeti]

In many auctions void showing specifically, 2N limit or better with singletons.

[jillybean]

If you have shortness or a void you must have a second suit unless 4441, why not bid a 2/1?

[smerriman]

This doesn't make any sense to me - why do you need two different bids to show weak hands? The reason for jumping to game when weak is to preempt the opponents; surely this just helps them?

There are definitely plenty of ways to show gf hands without a huge jump. Splinters are therefore for the lower range hands where the splinter is all that partner will need to know to make the right decision about looking for slam or not.

There's no way to say 'I want to look for slam if you have no wastage in this suit; honors in all other suits are suddenly worth much more' by starting with a 2/1 or forcing 2NT. 2/1 would imply your second suit is more important, and 2NT is when you need information from partner, rather than vice versa.

[johnu]

I play splinters as approximately 3 cover cards (~10-12 HCP) plus shortness. If you play a lot weaker, opener will almost never have a hand worth a slam try so you are leaking information with almost no return.

I also don't use 3NT to show a balanced NT hand, so I use 3NT over a 1♠ opening (3♠ after a 1♥ opening) as an unspecified singleton splinter (next step asks which suit), and direct double jumps as voids. The reasoning is that a well placed void is much more likely to result in slam than a singleton, and most hands opener won't have a slam range hand so you minimize information leakage.

[Stephen Tu]

I think bidding 1M-4m to show what other people simply bid 1M-4M would be a net loser. You give opps either room to stick a suit in under your 4M, or make a light takeout double to compete without committing to doubling your 4M, or let them show a 2 suiter by cue bidding your 4M, rather than limiting them to pass/double/bidding > 4M. The slams you could reach when partner has a moose with no wastage would be too infrequent compared to the losses in my estimation.

As for the splinters, what I play depends on how much system a particular partner will tolerate. Tiered splinters so that you can show multiple strength ranges, or perhaps alternately distinguish singletons from voids, I think are pretty useful. If you structure your responses to 1M-2nt well you can also put in additional splinters that were too strong for the direct option (e.g. I use 1M-2nt-3c! (min)-4x as spl too strong for 1M-4x). Normally I like to play the direct range as a good 11-14 hcp or so, weaker/stronger in other bids.

As for why you would want to splinter rather than 2/1 or use some forcing major raise like J2nt:-the first question is whether you want to show partner what you have, or query what partner has, which steers you toward J2nt/equiv or not.

The second question is whether 2/1 sequence is going to be a better description of your hand than just splintering directly. Do you really want to 2/1 on CKJxx/CQxxx, bid 3M, then bid an ambiguous 4d, and partner has no idea whether 4d is shortness or high card or how good your club suit is, or do you want to reserve the 2/1 sequences for hands with *good suits* that provide 4/5+ tricks if partner can help you out by cue-bidding an honor in it, that can maybe make slam even if partner has some wastage in your spl suit, if he has that key honor in your suit. If you have scattered but gf values, mediocre side suits, and a stiff, why not show it and reach those lower hcp slams when partner fits well? 2/1 won't do it because of the ambiguity later in the auction, partner can't distinguish the good side suit hands from the mediocre side suit hands + shortness if you don't have normal splinters available.

So if I have something resembling a prototypical splinter hand in the range I've agreed, I will splinter while I have the chance, to give partner the most accurate picture of my hand before the chance goes away, then abide by partner's judgment (he can also pass the buck back with like a last train cue if available). I will choose something else if I am too strong for the range, or have a GOOD side suit I'd rather show (on a good day being able to splinter later).

[mikeh]

The simple version of my two approaches is:

Splinters show roughly 11-13 hcp...from a sound limit raise to a minimum game force raise.

Over 1M, we use 3N as the hand that is too strong to jump to 4M but too weak to splinter. After 1S, 3N might be Qxxx x Axxxx Qxx

4H might be Qxxx x AJxx Kxxx

4S might be Qxxxx xx Kxxx xx

Note that I generally don’t splinter into the suit below the major, unless I can avoid it, since partner has bidding space between the splinter and 4M. This issue is reduced in significance by playing a narrow range for the splinter.


The more complex method is to use a jump in the other major as ‘any splinter’

This has two advantages.

1S 3H or 1H 3S shows about 11-13 hcp with 4+ support and a stiff somewhere

Opener simply bids game if he has no slam interest, which is a common situation, and the opps don’t know where the shortness is. A common defence to splinters is, when not vulnerable, to use double to suggest saving in that suit. So this approach usually deprives them of that ability.

If he has interest, he bids the cheapest step. Now responder identifies his shortness, using step responses

1S 3H 3S: 3N is clubs, 4C is diamonds, etc. Notice the saving of bidding space.

Another advantage is that we can use 4 level new suits as void showing. We can’t show a spade void over 1H, but we can show all other void splinters over 1M. There is a profound difference, in terms of slam bidding, between a void and a singleton, when opener lacks the ace of that suit.

As for other ways to raise with good hands, as Stephen noted, whether you use 2/1 in your own suit, or splinter, or use J2N, depends on whether your hand is best suited to describing itself or to taking control.

I won’t use a forcing raise with my own decent 5 card suit...I want partner to later know that Qx/Kx/Ax are very useful holdings in my suit...they represent a source of tricks. Note that on some auctions, one gets to 2/1 and still splinter, especially if 2M is the default rebid (as I always play). 1S 2D 2S 4C would be primary spade support (might be Hxx on this auction), 5 diamonds, a club stiff and an opening hand.

Generally do NOT use splinters on powerful hands. You need to be able to respect opener’s sign off.

[Zelandakh]

Typically a normal splinter shows a minimum game force or a hand worth an additional try and a maxi-splinter shows the range in-between. Opposite an opening bid that might equate to 13-15 or 20+TP and 16-19TP. There are also cases where a splinter shows more than a minimum game force, with an obvious case being opposite a 2♣ opening. It can also be quite useful to separate out singleton and void splinters. Now a void splinter might look not so different from your weak splinters, since 13-15TP becomes something like 8-10hcp. As for why to splinter rather than using a delayed raise sequence, in 2/1 you want this to show a good suit so that Opener can correctly evaluate fitting honours. The ability to do this is one of the key advantages that 2/1 offers over some other systems so you really do not want to be diluting that and forcing its use on poor suits.

In competition, splinters and their cousins fit jumps are often used somewhat differently from purely constructive auctions. Quite often the aim here is as much to get the level right as for investigating slam. This is particularly true after a nebulous 2-suited overcall where the side that finds out whether there is a double fit tends to have a decisive advantage. Vulnerability matters here and it pays to have clear rules about which levels and which vulnerabilities create forcing passes, since this has a direct effect on how the calls can be used.

[jillybean]

Thanks again for the detailed responses, now I have something more to discuss with my partners.

Q. In regards to splinters, what do you do with a singleton Ace?

[Winstonm]

jillybean, on 2021-April-07, 13:53, said:

If you have shortness or a void you must have a second suit unless 4441, why not bid a 2/1?

You can have 5 or 6 card trump support, don’t forget 🤔

[smerriman]

jillybean, on 2021-April-07, 20:51, said:

Q. In regards to splinters, what do you do with a singleton Ace?

I've seen a lot of people saying you shouldn't splinter with a singleton Ace, because that will cause partner to misevaluate holdings like KQx as useless.

That logic has never made much sense to me; while it makes sense for that suit in isolation, the chances are KQx vs a stiff Ace is going to help you find a slam that's light on HCP strength is pretty low.

To me it's more that if you have a singleton Ace and the normal splinter strength, you're highly lacking in values elsewhere. You're probably going to be more interested in whether partner has shortness in your weakest suit - ie starting with 2NT - than vice versa. But I wouldn't rule it out entirely; sometimes a splinter is the still the least of evils.

[DavidKok]

I'm a bit late to the party, others have already made some very good suggestions. I'd like to add that, from a system design perspective, it is crucial that your bids that take up 2-2.5 levels of bidding space are narrowly defined. Mikeh mentions 11-13, my range is slightly wider and lower at 9-12 points/7 losers (did I mention I like to bid aggressively? The shortness is supposed to make up for the lack of points, and this range increases the frequency of the bid just a tiny nudge) but the idea is the same. Also it is standard to include some monster hands that are prepared to play on the 5-level even over a signoff, so the real range is split 9-12/17+. I've only seen this happen once, and that was a hand I was defending.

So what do you do with hands outside this range, a good trump fit and shortage? If you are stronger you can start with a 2/1 auction (typically 2♣). This does mean your 2/1 bids lose a bit of definition, but usually you can catch up on the second round. With weaker hands there are two options - you can systematically reserve a bid for these hands (such as 3M or 3NT), or suppress the shortness and give a simple or limit raise (and hope partner will inquire further). I think both approaches are very playable. Reserving 3NT for a serious splinter (let's say 13-16, going off my numbers above) is also gaining in popularity despite the overlap with powerful 2/1 auctions.

I personally don't distinguish between a void and a singleton. I think it does not happen often enough that this distinction makes or breaks the slam, so I don't much like reserving special bids for them. A splinter has the primary goal of informing partner that their points are or are not 'working' and if we're playing in a '30 or 34 point deck', and that's largely the same opposite a singleton or a void. Put differently, partner will likely respond the same way over a singleton and over a void, so it makes sense to merge the two into a single bid.

Lastly regarding splintering with an honour (you mention the ace, personally I'd also include the king and depending on the preceding auction the queen may also be relevant), I don't like it. I agree fully with smerriman's explanation. Keep in mind the goal of splinters is to find sharp slams in 30 (or 34) point decks - where the opponents have concentrated values in a suit that will be ruffed. This doesn't work if you have strength in your short suit. It also helps that on a non-splinter slam try auction you can simply show the singleton ace or king as a control, so the splinter is often not even the smallest lie.

[Winstonm]

I'll basically restate my comment from a previous thread: IMO, the only purpose of a splinter is to find a slam. My old partner called it a search for a 30-point deck, meaning that x opposite xxx leaves one loser and all other cards working. With the 30-point deck in mind, 14 opposite 14 becomes slam range, less if there is a secondary suit that runs. Based on this, a splinter around 11+ to 14 seemed right to me.

[mycroft]

I'm not going to say much new, because it's all by others. But maybe I'm saying it differently.

One thing I heard that made things make sense to me: "when you bid 2NT (GF raise), you're captain (for now). When you splinter, partner is captain." Thus, the range on splinters is limited - "you have to be able to respect partner's signoff". More limited the higher the splinter - 4M-1 needs to be extremely narrow, because partner has no room to consider.

"Standard" responses to J2NT are "partner, please splinter". The intent is the same - to see how the hands fit; but again "You're captain". More detailed responses have a "warning, minimum" response, to avoid getting out of control (or giving away too much with minimum opposite minimum), but still, if slam investigation happens, relays into "please splinter".

Mini-splinters (single jump shifts) are common in my area. I don't mind that (at least it stops people from WJSing into 3♣!), especially if rationally combined with double-jump splinters, giving three ranges:

• 8-10ish with shortness: mini-splinter
  
• 11-13ish: full splinter
  
• 14"plus": mini-splinter again, but override partner's decision.

Like many, I prefer showing first and second round control with A-singleton over splintering - partner will see QJTx or KJTx as wasted values and downgrade their hand, when in fact they're pulling their full weight. Probably same with stiff K - again, it shores up "wasted" honours in partner's hand, and Axxx isn't quite the powerhouse it is opposite x-and-King-elsewhere. But all of that comes from the "I'm describing my hand to partner so they can make the decision, am I giving them the right picture?" than any fixed rules.

[Zelandakh]

jillybean, on 2021-April-07, 20:51, said:

Q. In regards to splinters, what do you do with a singleton Ace?

I would strongly suggest not splintering with a singleton ace unless you have absolutely no other way of showing your hand. Most of the time it causes partner to misevaluate their hand, which is precisely the opposite of what you want from a splinter auction.