I believe there are some older discussions about this topic on tiltforums, but it can’t hurt to remind folks where this system originated from, and how it’s been used and misused since.
Way back in the olden days, PAPA used to do it’s ticket scoring based on the total points scored on a ticket, which seems insane today, but was “state of the art” for tournaments back in the late 90s. After some awesome discussions on rec.games.pinball, Kevin Martin gave his wacky (at the time) idea of ranking-points-based-on-position a try at PAPA 6 in the “Champions Division” (although the main division of PAPA 6 still went with total points). That was in 1998, and Kevin used his new system for the old Pinburghs from 1999 through 2003, and then when PAPA 7 came back in 2004. So it was Kevin Martin’s idea to give a 10 point “bonus” to the highest score and a 5 point “bonus” to the second highest score via 100-90-85-84-83…
The important takeaway here, though, is that these bonus points were designed for a 5 game ticket system. And since only your best ticket counts in the standings, the chances that a #1 or #2 score on a game for the entire tournament occurred on someone’s best ticket was just not that likely (for instance, at PAPA 18, only 4 out of 24 qualifiers in the “A” division had a #1 or #2 score on their qualifying ticket). So it was never a huge deal at PAPA; just a fun oddity that always gave Belsito a chance on his final run 
So really pump-n-dump tournaments should never have kept the bonus point structure in scoring, because unlike ticket based qualifying, every game in a best-game format counts, so there was always going to be a #1 and #2 score influencing the rankings (At PAPA 20’s pump-n-dump qualifying, all the top 12 had at least one #1 or #2, and 16 out of the top 24). But even then it wasn’t too big a deal because at first, pump-n-dump tournaments required you to get a best score on every game in the tournament. It really wasn’t until the PAPA Circuit took off years later that you had tournaments where they needed to have a bank of games much larger than the number of games required to qualify on because of the sheer number of players wanting to play. And again, instead of thinking it through, the 100-90-85-84-83… scoring system just came along for the ride. It was at one of these tournaments (an old Expo, I think) where it was like your best 5 games out of a bank of 12 games, and since there were now going to be 12 games where someone got the 10 and 5 point “bonuses” there was a real chance that someone could have 5 scores in the top 16 (without a first or second), and not come close to qualifying in the top 16! With all of those bonus points polluting the water, you suddenly needed to average a top 8 or 9 score on all 5 games to have a chance at qualifying! Which seems insane for the format, really.
So it was at that point, that TDs started to temper the bonus points down from 10 and 5 to 5 and 3, or remove them all together. Personally, having a bonus for the top score seems just fine to me, but having a bonus for the #2 score just seems arbitrary and silly. And really, I believe the very best scoring ladder when ranking pinball scores is to use a constant function where Score N+1 is P percentage of score N until it gets down to 0… I posted some samples way back when, but something like:
[p=.97] [p=.95] [p=.93]
100 100 100
97 95 93
94 90 86
91 86 80
89 81 75
86 77 70
83 74 65
81 70 60
78 66 56
76 63 52
74 60 48
72 57 45
69 54 42
67 51 39
65 49 36
63 46 34
61 44 31
60 42 29
58 40 27
56 38 25
54 36 23
53 34 22
51 32 20
50 31 19
48 29 18
47 28 16
45 26 15
44 25 14
43 24 13
41 23 12
40 21 11
39 20 11
38 19 10
37 18 9
36 17 8
34 17 8
33 16 7
32 15 7
31 14 6
30 14 6
30 13 5
29 12 5
28 12 5
27 11 4
26 10 4
25 10 4
25 9 4
24 9 3
23 9 3
22 8 3
22 8 3
21 7 2
21 7 2
20 7 2
19 6 2
19 6 2
18 6 2
18 5 2
17 5 1
17 5 1
16 5 1
16 4 1
15 4 1
15 4 1
14 4 1
14 4 1
13 3 1
13 3 1
13 3 1
12 3 1
12 3 1
12 3 1
11 2 1
11 2 1
10 2 0
10 2 0
The thing I really like about these scales is that I think it correctly reflects the pinball scores themselves in that once you get down to 50th or 75th place… A score of 7,239,480 on say, Cirqus Voltaire is basically equivalent to a score of 7,237,940, even though one is a smidge higher than the other… Both should get the same number of ranking points, IMO. I believe I have seen some European tournaments used this style of ranking points and it always made the most sense to me, but I understand that it could be confusing to someone unfamiliar with all the nuances involved in ranking pinball scores.
The other advantage to using a simple degrade function is that it will work consistently on any starting value (100, 150, 200, etc…) depending on how many competitors you have. And, really, that’s the other half of this issue that is super important - Your scoring scale absolutely needs to be appropriate for your expected number of competitors. 100,90,85,84,83… was becoming silly at PAPA because it went down to 0 too soon. At PAPA 18, CFTBL was played 264 times! and the 88th best score of 113,136,280 was worth exactly the same number of ranking points (zero) as the 264th best score of 2,686,980. By only giving points to essentially the top 30% of scores, PAPA qualifying put a much higher premium on scoring 3 big scores on a ticket, instead of having 5 solid above-average scores… INDISC this past year used 200,190,185,184… so that the top 180 scores or so scored some points and that really put a premium on having a consistent, 5-game ticket instead of hitting 3 home runs with 2 strike outs.
As always, YMMV.