'test meld' suggestions

[tony ennis]

On the meld test page, I suggest 'test weight' be explained and the 'elapsed seconds' rounded to the nearest hundredth.

[rakbeater]

AGREE!!!!

[mickmackusa]

(04-28-2013, 10:49 AM)mickmackusa Wrote:  Goal: Perfectly count the meld of 10 randomly generated hands in under 5 minutes.
Statistics: Seconds Elapsed is measured from click of the Start button to the click of the Finish button and is recorded to 14 decimal places. Test Weight* is the measure of how many total meld units (marriages, pinochles, arounds, runs) need to be found.
Scoring: round_to_nearest_integer( ( 300 - "Seconds Elapsed" ) x 10,000 x "Test Weight" )

Upon completion of the Meld Test, the member will see:
1. the correct meld counts for each hand.
2. their accuracy, time (if accuracy is perfect), and score (if time is less than 300)
3. a notification if any personal or site high scores have been surpassed.

*Test Weight is "compensated" 3 additional points for each roundhouse when the selected scorechart designates a point value for roundhouses (currently only NPAPinochle.org). This is done so that no chart has an unfair advantage/disadvantage regarding speed and scoring. Given there is only 1 roundhouse in a 10-hand game:
- NPAPinochle.org chart will score 4 points (roundhouse[1] + compensation[3]).
- All other charts will score 6 points (trump marriage[1] + three non-trump marriages[3] + kings around[1] + queens around[1]).
The ease of counting designated roundhouses (automatic 24 points) becomes clear when additionally calculating runs, multiple kings around, or multiple queens around; all of which share cards with the roundhouse.

I don't want to overload the Meld Test page with too much text.
The above quote from my explanation post could be more than the average player wants to read.
They might just want to play the game without having to scroll down.
Is there something specific from my explanation post that you feel is necessary? Or a more concise way of wording?
I am open to suggestions.

[ToreadorElder]

They list the score for a roundhouse not because it's a scoring unit, but because they're fairly common, and it's an easy shorthand to remember. Throw out 'roundhouse' as a unit. It'll count the same, and it'll make each score sheet give the same results. You don't have that now; you are giving the NPA chart a definite disadvantage.

[tony ennis]

mickmack's explanation suits me fine. Thanks!

[rakbeater]

(07-08-2013, 02:53 PM)ToreadorElder Wrote:  They list the score for a roundhouse not because it's a scoring unit, but because they're fairly common, and it's an easy shorthand to remember. Throw out 'roundhouse' as a unit. It'll count the same, and it'll make each score sheet give the same results. You don't have that now; you are giving the NPA chart a definite disadvantage.

Says the guy who hasn't played the game once. What are you waiting for?

[mickmackusa]

(07-08-2013, 02:53 PM)ToreadorElder Wrote:  They list the score for a roundhouse not because it's a scoring unit, but because they're fairly common, and it's an easy shorthand to remember. Throw out 'roundhouse' as a unit. It'll count the same, and it'll make each score sheet give the same results. You don't have that now; you are giving the NPA chart a definite disadvantage.

I feel that it is my duty to defend my program. While I agree that NPAPinochle.org's recognition of the roundhouse is an easy shorthand for counting a common combination of cards, I disagree that the end count will be the same. (As a note, my family plays roundhouses.)

As I mentioned in my explanation post, meld count discrepancies occur when a roundhouse shares cards with a run, multiple K's, and multiple Q's.

Examples:
1. Roundhouse + Run (click this permalink to the Meld page)

-- if you toggle the Scorechart Selector from Yahoo.com to NPAPinochle.org you will find that the score is different and not just because the Run is worth 10 more points at NPA. ...if Runs at all sites were worth 15, then this hand at NPA would be 39 points, and Yahoo would score this 35. The NPA roundhouse value includes the assumed royal marriage points even though there is a run.

2. Roundhouse + Multiple K's (click this permalink to the Meld page)

-- Yahoo.com scores this 96 meld. NPAPinochle.com scores this 104 meld. The NPA roundhouse awards the single kings around value (+8 points) despite the occurrence of double kings around.

3. Roundhouse + Multiple Q's (as it acts the same way as #2, I won't list the scoring difference)

Programming the Meld page took me a long time, as the members may recall the whispers about a new program coming and it didn't actually roll out until several months after. If NPA's scorechart was like the rest, my job would have been simpler and faster. However, it is how it is, and it's not my decision to change an organization's scorechart.

Rakbeater and I carefully discussed how to compensate for the differences in scorecharts. Without compensation, the NPA chart would test faster as there are less units to find/add up, but we felt that the Test Weight given to all the other charts would have been too great. We have no special allegiance to NPA, we are only trying to be fair and just.

IMO, the only arguable aspect is the arbitrary compensation value of "3" that we have adopted. Maybe some people will say that it should only be "2" or perhaps "4", but I think then it just comes down to personal opinion.

[ToreadorElder]

<sigh>

Turns out, you're probably right.

http://www.pagat.com/marriage/pindd.html

Scroll down a bit more than half the page, to the variations in scoring. They show 'pinochle 14' which is close enough to what NPA does, and they show the roundhouse as a separate scoring unit. So, my bad.

But IMO they're grossly inflating meld...probably at least 15 points per hand. Roundhouses aren't all that common, but one can largely assume a run almost every hand, and I'd guess a pinochle every 3rd hand...that's enough to account for +15 points on average, in total meld at the table. And the competitive swings are larger...hands where each side has a decent run. The swing's 46 points instead of 26. This has enormous tactical implications.

[rakbeater]

(07-09-2013, 04:44 PM)ToreadorElder Wrote:  <sigh>

Turns out, you're probably right.

http://www.pagat.com/marriage/pindd.html

Scroll down a bit more than half the page, to the variations in scoring. They show 'pinochle 14' which is close enough to what NPA does, and they show the roundhouse as a separate scoring unit. So, my bad.

But IMO they're grossly inflating meld...probably at least 15 points per hand. Roundhouses aren't all that common, but one can largely assume a run almost every hand, and I'd guess a pinochle every 3rd hand...that's enough to account for +15 points on average, in total meld at the table. And the competitive swings are larger...hands where each side has a decent run. The swing's 46 points instead of 26. This has enormous tactical implications.

Part of the reason why things take so long to come to fruition here on the site (besides the obvious time restraints, since this is all volunteer and non-paid) is that we are really particular about getting things right. We have done a lot of behind-the-scenes homework and research, and continue to do so. We try really hard here. (Glossary page - I know I know)

[ToreadorElder]

My issue was simply the whole double-counting thing. It's not spelled out very well, that this was the intent, and it's uncommon to do this. I hate bad, confusing rules.