(An attempt at) The top 40 TPA players from leaderboard scores

[Bowflex]

I think we need a ranking of best usernames. With all due respect to everyone else, I kind of like Moldygoat.

 

[Slam23]

Nice list, I see myself at a modest 32nd place, with which I am content! I usually lose interest in playing a table beyond the required point total for 1000 HOF points on mobile (which in my opinion is the harder platform to play on because of simplified nudging controls) Although I really respect the stamina and concentration level that is required to get in the top tier scores, I simply lack the playing time to rack those up. Not to sound sour about it though, because if I would have the time I probably would lack the concentration skillz! So hats off to the Viking!

 

[nstalkie]

Any chance of publishing a full ranking somewhere ? maybe where you can update it regularly ?
I'm wondering where I'm placing, as I have 3 top 100 scores I would have had 4 but apparently my star trek score is only enough for 101th place right now (nstalkie)

Still wish we had separate leaderboards + a leaderboard reset.

 

[andrewduhan]

I ran through my data to get a top 100 players list based on the top 2000 scores on each table (instead of top 100).
Scoring is the same as OP.

Rank	Player	Points
1​	maestroreese​	81881​
2​	dylan_h​	81317​
3​	pinballchris​	80974​
4​	Crusty Booger​	80309​
5​	Space Tractor​	79789​
6​	invitro​	78458​
7​	vikingerik​	77279​
8​	DeeEff​	76133​
9​	tinder​	76029​
10​	MT​	75169​
11​	Nightmac21​	75131​
12​	Jazza​	75026​
13​	geohoo​	73808​
14​	ChristofferB​	73613​
15​	AntonR​	73404​
16​	atoppachar​	73277​
17​	slam23​	72252​
18​	-SVB-​	72071​
19​	TRLBrony​	70960​
20​	Blah​	70824​
21​	victorstulemeije​	70303​
22​	TOUGE_KILLER​	69755​
23​	Tripple I​	68824​
24​	Fungi​	68298​
25​	RagingBull888​	67674​
26​	kosacks​	67525​
27​	gooche77​	66411​
28​	Shooting Shark​	66260​
29​	PieCES​	66005​
30​	Peter Williams​	65238​
31​	Inspector42​	64873​
32​	CheesyBeefy88​	64872​
33​	Smash Crate​	64438​
34​	rehtroboi​	64343​
35​	Zoop25​	64109​
36​	ER777​	64027​
37​	kan​	63984​
38​	Vic Golf Chad​	63892​
39​	FALCO12-QUEBEC​	63110​
40​	Sir_Moovenstein​	62939​
41​	rockman_x_2002​	62591​
42​	Chris​	62268​
43​	chk​	61855​
44​	ppeltola768​	61770​
45​	ryan​	61697​
46​	xNiCeGuYx​	61468​
47​	LUK​	61439​
48​	mjohns​	61316​
49​	pinballwiz45b​	60584​
50​	torstenkuchenbae​	60426​
51​	RZR​	60197​
52​	Crazy Newt​	59683​
53​	pincusson​	59586​
54​	Toolinit​	59543​
55​	Cutepuppy​	58777​
56​	LordEngi​	58695​
57​	djchrille​	58550​
58​	Badasstronaut​	58549​
59​	clawhammer​	58513​
60​	9u1d0​	58200​
61​	nonojin​	58087​
62​	nikostos44​	58070​
63​	HUFMUF​	57732​
64​	Krakehz​	57325​
65​	aspiratatator​	56863​
66​	Mr-Bouzouki​	56451​
67​	Buzzed Light​	56383​
68​	SoonPoker​	56373​
69​	PrimeroIncognito​	56264​
70​	monst​	55910​
71​	ksmmspt​	55791​
72​	Pezpunk​	55338​
73​	mrgamesmaster​	55333​
74​	Acctarheel​	54529​
75​	hrr​	54431​
76​	djag29​	53678​
77​	CarShark​	53617​
78​	flipmon​	53384​
79​	Carl​	53281​
80​	nowclickthis​	53005​
81​	johnny​	52958​
82​	Lordbobby​	52633​
83​	Kaoru​	52613​
84​	DonMagnuso​	52075​
85​	Tensikka​	52026​
86​	tiotbellot​	51652​
87​	mertzee​	51647​
88​	alx002​	51342​
89​	SHO​	51190​
90​	comboace​	51153​
91​	moldygoat​	51000​
92​	karl​	50999​
93​	Pgh-Navigator​	50916​
94​	steven_dutch​	50782​
95​	Jake Alt​	50733​
96​	Nico​	50637​
97​	kay989​	50615​
98​	cdub​	50382​
99​	harry82988​	50220​
100​	logix​	50191​

[thead]

[/thead]
[tbody]
[/tbody]

 

[invitro]

I scrape the top 2000 for every table every few weeks or so.

Amazing stuff! Is there any chance that you can post the scraping code... or email it to me? Please? I cannot figure out how to do it, and would love to be able to! Thanks for posting your stuff, anyway!

 

[Carl Spiby]

Woo 79.

 

[Pinetwig]

Awesome job! I rank better than I expected.

 

[Espy]

Not in the top 100, though I expected that as there's a lot of older tables I haven't revisited since I got good at the game. For the most recent tables, most if not all of my scores have been top 100 (with my best being #5 on El Dorado).

 

[andrewduhan]

Amazing stuff! Is there any chance that you can post the scraping code... or email it to me? Please? I cannot figure out how to do it, and would love to be able to! Thanks for posting your stuff, anyway!

How I scrape is no secret or magic, but I don't want to post the code.
I'm just scraping occasionally for my own personal interest, but FarSight might not like it if lots of people started doing it too. I haven't found a way to fetch more than 10 scores at a time, so pulling 2000 scores for 42 tables means 8,400 individual requests. It takes almost four hours to pull it all down because I spread out the requests to not flood their server.

It would be awesome if FarSight would make an API available so we could fetch scores in much bigger chunks. If they can't or won't, I can make the score data available publicly - but I wouldn't do that without FarSight's consent (it is their data after all). This is why I haven't put my charts in a public place - they are interactive and searchable and that means the data is very easy to extract.

Anyone at FarSight reading this topic? With your blessing, I'd love to share the charts & data. I could also help you with scoredata API design and implementation if this is interesting to you but you don't have the resources.

-Andrew

 

[invitro]

How I scrape is no secret or magic, but I don't want to post the code.

Ok... you gave me the hint I needed to finally figure it out for myself . Fantastic! I didn't think it could be done but I should've have given up so easily. All that hand copying for nothing but... fun?

Hmm... now what, I don't think I will grab that many scores, the top 100 seems enough for me.

With easy access to data, I may post the rankings regularly, if there is interest. Unless you do! I do think scoring only the top 100 scores on a table is a better way than using all scores... I obviously am not #7 or #8 but #25 seems possible, unless that is too high also. If anyone has a better way of ranking players then by all means share.

Have you done any automated searching for hacked/glitched scores? Just curious if it's worthwhile... the bad scores in the top 10 can easily be seen just from the website, but I don't think I would've noticed the RBioN scores unless I had gone through all of them by hand.

Have you attempted to model the score distributions? That might be of a decent amount of interest.

I know I would be very interested if things happen such that you make the scores and your graphs and charts publically available.

 

[andrewduhan]

graphs for leader data on all tables

I would love to scrape more often and post the graphs & data.
I'm grabbing the top 2000 because a player who is, say, 500th in all 44 tables would wind up ranked about 30th overall. Scraping only the top 100 scores ignores a strong-but-not-top-level player.

Here's a link to a (large) screenshot of the graphs for all 44 current tables.

Data is current as of yesterday. Things I notice:

• The "billionaire bumps" on BOP
• Weirdness at the top of Genie
• Some wildly high scores on Big Shot, BK2000, Central Park, Dr Dude, Haunted House, TOTAN, and Tee'd Off
• Previously mentioned glitchiness on RBION

I think you can also get a sense of which tables reward strategy or can be 'infinite' (steep curve, pros hitting large rewards or lots of extra balls) and which require consistent skill (shallower curve, longer playtime required)

 

[vikingerik]

Bride's billionaire bumps are hilarious, awesome visualization! El Dorado has similar lumpiness because bonus is so dominant, the score is heavily quantized around the number of EBs earned.

Dr Dude is legit I'm sure, Tarek Oberdieck is a great player, has been for years on the Pro Pinball leaderboards.

Genie is because of the 3M wizard goal, players play until they reach that then stop. Taxi has a similar lump at 27M which is the score for 1000 PHOF points.

Pinbot has noticeable jumps around 12M and 22M. That's because Pinbot awards another replay (EB) each time you reach the replay score after a rollover, so every time rolling over past x2,000,000 lets you go a little longer.

CFTBL also has some kinks in the curve. This table has trouble detecting rollovers from extremely large super jackpots (like 1.5B x4) which could be distorting the data.

NGG and STTNG I would have expected to stand out more thanks to their highly progressive scoring nature, but they both look pretty normal. I guess only the top few dozen players have had the hourlong games to see where it really starts inflating.

 

[Pinetwig]

Genie is because of the 3M wizard goal, players play until they reach that then stop. Taxi has a similar lump at 27M which is the score for 1000 PHOF points.

It's 40M for 1K HoF-points on Taxi.

 

[andrewduhan]

Nice notes, vikingerik. Exactly the kind of insight I hoped to gain from these charts.

 

[Slam23]

This 17th place seems to reflect my playing style more, and is more near to my 6th place on the mobile HOF leaderboard. I tend to buy all tables, and mostly score at least quite decent on them, on some of them I reach higher scores but I often come up short of the top 10-20. There are some incredible scores on the leaderboards (not counting the obvious bogus scores), some of which must take a whole day of playing, especially on tables that don't have an exponential scoring pattern. Nice graphs! This does give insight and I heartily concur with VikingErik on the explanations. I would venture that maybe people are somewhat afraid of a rollover after 1000B on STTNG? I don't now if the ROM can handle that kind of scores. Those Y-axis numbers for TOTAN pose somewhat of a challenge to us players

 

[PinballLizard (Mark Wyda)]

For those interested in further statistical exploration of pinball scoring from a slightly different perspective: Why You Don't Achieve At Least Half Your High Score At Least Half the Time

Also, RBION is a really bad table to use as an example, as there was a bug early on in its life that added 10B (and in a few cases 20B) to the actual scores. That's why you have that huge concentration of scores at 10B interrupting what would otherwise be a nice exponential curve.

I read the article, and I have to say I disagree with it.

I can only speak for myself but, my best 5-7 tables and best scores aren't really "1 in 1000", lucky games. Nor do I score big, only to "frustratingly" get crappy games afterwards. It doesn't work like that with me. Every time I go back to a table I'm good at, I pretty much get my best or at least close to it, every time (as long as I'm trying and not messing around). My top 5 is all within fractions of each other, so to put it in generic terms, if my best was 140 million my top 5 is 140,139,128,135,130.

Now there are breakthrough games, where I've blown my previous best out of the water by a substantial amount, but then that becomes the new norm, because in order to have that breakthrough my brain has adapted, I've become more comfortable with a table, learned a trick to avoid drains, not made "stupid" plays, came up with a better strat , what have you.

Now sometimes I'll have that insane ball 1 that is super close to my all time score, and think holy **** I'm about to blow my old score out of the water... Only to house ball my way back and end up only beating it by a small margin. But that's more psychological than anything, I call that psyching myself out and playing poor pinball as a result.

I guess you could kinda call that the 1-1000, but a ball that good is also common for me. The way I play is I don't pay attention to what ball or how many balls I have, because playing correctly in a sense means you can go forever. This is especially true of pinball arcade , digital where randomization is less than real, things behave in a set way every time, and the way nudging works in digital. So what I'm saying is that, every time I drain it's a mistake I made and not "the games fault", and theoretically I should be able to do it all on one ball, all the subsequent balls are just "do over" buttons for when you screw up

Having a really good score and then scoring substantially lower every subsequent try sounds like me when I first got the game, and wasn't very good. I'm not degrading you or anyone, all I'm saying is that once you get over a certain hump , you kinda control your own destiny after that. You can still plateau, sure, but that's what it is a consistent plateau vs a curve.

 

[Espy]

Great job Andrew, these are really insightful. And it's funny how much the overall trend can be linked to player behaviour (such as Genie: "I made 3M! I can relax now... aw, I lost the ball."), or Glitches such as Ripleys.

 

[invitro]

Thanks lots for the graphs a.d. and notes v.e.! Very interesting stuff.

 

[Sean DonCarlos]

I read the article, and I have to say I disagree with it...

So as not to derail this thread, and not to distract from andrewduhan's excellent graphs, I've copied your comment into that blog article and responded there.

 

[vikingerik]

It's 40M for 1K HoF-points on Taxi.

Hmm, ok. I play on PC which doesn't have the HOF goals so was going from memory.

I remember chasing 27M on Taxi for some reason but now can't remember why. Replay is much lower than that and there isn't any score wizard goal, right?