Bad explanation - splinter when supporting

[thorvald]

[hv=url=https://www.bridgebase.com/tools/handviewer.html?lin=st||pn|Human,~~v3fakebot,~~v3fakebot,~~v3fakebot|md|2SAHAJ92DKQT94CK87,SQJT843H75DJCT942,SK965HKT8DA865CJ6,S72HQ643D732CAQ53|sv|b|rh||ah|Board%204|mb|P|mb|P|mb|P|mb|1D|an|Minor%20suit%20opening%20--%203+%20!D;%2011-21%20HCP;%2012-22%20total%20points|mb|P|mb|1S|an|One%20over%20one%20--%204+%20!S;%2011-%20HCP;%206-12%20total%20points|mb|P|mb|2H|an|Opener%20reverse%20--%205+%20!D;%204+%20!H;%203-%20!S;%2021-%20HCP;%2018-22%20total%20points|mb|P|mb|4D!|an|Splinter%20--%204+%20!D;%2010-12%20HCP;%20forcing%20to%205D|mb|P|mb|4N|an|Blackwood%20(D)%20--%205+%20!D;%204+%20!H;%203-%20!S;%2021-%20HCP;%2022%20total%20points;%20forcing%20to%205D|mb|P|mb|5D|an|One%20or%20four%20key%20cards%20--%204+%20!D;%2010-12%20HCP|mb|P|mb|6D|an|5+%20!D;%204+%20!H;%203-%20!S;%2021-%20HCP;%2022%20total%20points|mb|P|mb|P|mb|P|pc|H7|pc|HT|pc|HQ|pc|HA|pc|DQ|pc|DJ|pc|D5|pc|D2|pc|D9|pc|S4|pc|D6|pc|D7|pc|DT|pc|S3|pc|DA|pc|D3|pc|S5|pc|S7|pc|SA|pc|S8|pc|H2|pc|H5|pc|HK|pc|H6|pc|SK|pc|S2|pc|C7|pc|SQ|pc|S6|pc|C5|pc|D4|pc|SJ|pc|HJ|pc|C4|pc|H8|pc|H4|pc|H9|pc|C2|pc|C6|pc|H3|pc|CK|pc|CT|pc|CJ|pc|CA|pc|C3|pc|C8|pc|C9|pc|D8|pc|S9|pc|CQ|pc|DK|pc|ST|]399|300[/hv]

 

Note the alert and explanation of 4♦

[smerriman]

Looks like one glaring mistake in the ruleset. For passed hands, its rules about 4 level bids are:

 

1) jump to 4 of opened suit with 11+, 4+ card support, no support for second suit, 3+ controls, describe it as a slam try

2) jump to 4 of opened suit with 9+, 4+ card support, any outside singleton/void, 3+ controls, describe it as a slam try

3,4) same as 1,2) but jumping to 4 of opener's second suit

5) jump to 4♦ if opener's second suit was hearts with 9+, 4+ card support, 0-1 diamonds, 3+ controls, describe it as splinter

 

5 was presumably meant to only apply in the 1♣ - 1♠ - 2♥ - 4♦ case, but they forgot to check that opener's first suit wasn't diamonds, so the two rules overlapped. The result is that it bids 4♦ in both the splinter and non-splinter case, but alerts it as a splinter for both (it can't show two text alerts at the same time, and the splinter one came last, but the suit length part uses the first rule).

 

Trivial fix to implement.

 

Talking of reverses and splinters reminds me of my favourite bid description of all time. I should look into why that happened, since I posted that before I had debugging capabilities..

 

[edit] OK, it just defines 4x as a splinter over 3M for any x except the three suits bid previously, forgetting to exclude x=N, so it tried searching for N in the suit list, and overflowed into another part of the data.