◉The Wins Feed Open the partner account
paid
Affiliate disclosure. The partner link in the masthead and in the bands beside the copy on this page is a sponsored link to a partner operator, and this site may be paid if you open an account through it, at no extra cost to you. It carries rel="sponsored noopener" and opens in a new tab. That matters on this desk in particular: the subject is how much a wins display leaves out, and this site carries commercial marks of its own. No operator, game, studio, streamer or regulator is named, rated or recommended anywhere on this site.
The Wins Feed / The shape
Desk 65 - the wins you are shown

The bands the rows are taken from

A feed is not just a sample. It is a sample weighted towards the top of the range, because the top of the range is what a display is for. The weighting is measurable, which makes it the most useful thing a reader can learn from a feed.

Open the partner accountRead a feed in six minutes
Feed stub
Qualifying wins
457
Top band share
3.1% of wins
Top band in feed
50.0% of rows
Over-representation
16.3x
the barThe multiple a win must reach to be eligible for the display - on the sample, 250 times the stake
the poolThe players and spins the rows were drawn from - on the sample, 4,750 players placing 1,900,000 spins in the hour
the intervalThe hours between one player's own qualifying wins - on the sample, 10.4 hours at 400 spins an hour
Direct answerA wins feed is weighted towards its rarest band. On the sample, wins of 2,000 times or more are 3.1% of the 457 qualifying wins in the hour but 50.0% of the 12 rows shown - an over-representation of 16.3 times - because the display is drawn to show the extreme case rather than the typical one.

Two distributions of the same 457 wins

The first column is what happened; the second is what was displayed. Both describe the same hour, and the difference between them is the editorial rule of the feed.

The hour’s wins, by band, and the feed’s rowsband            wins   share    rows   share   ratio
250 - 500x       375   82.1%      2    16.7%   0.20x
500 - 2,000x      68   14.9%      4    33.3%   2.24x
2,000x and up     14    3.1%      6    50.0%  16.3x
total            457  100.0%     12   100.0%
                                        
mean multiple   the feed’s mean is dragged up by its top band
median multiple 328x across the wins, 1,640x across the rows = 5.0x

The rarest band is not merely present in the feed; it is more than half of it. The most common band, which is four fifths of the hour’s wins, is a sixth of the rows.

Why the weighting is rational

A display is not obliged to sample fairly, and there is a reason it does not. Twelve rows taken proportionally from the bands would be ten small wins, one medium one and a fragment of a large one - which is a dull display and, more to the point, an unrepresentative one for the thing a feed is selling.

What the feed optimises

A row that reads as large, a feed that reads as busy, and a case at the edge of the range so the top of the display is a number worth looking at. Every one of those is served by taking rows from the top band, and none of them is served by sampling the middle.

What the reader needs

The share of wins in each band and the interval between one’s own wins, which are the two numbers the display is arranged not to show. They are also the two numbers a reader can estimate from the feed itself, by counting bands rather than reading amounts.

Measuring a feed from its own rows

The bands are visible in the rows even when the shares are not, so a reader can do the comparison the display will not. The method needs nothing but the feed and five minutes of attention.