Tourney League Points Help (yes again) (1 Viewer)

[Mr Winberg]

My 2 cents:

I agree with all those that have said that the point system should encourage what you want from your players. If I ever started a league, I would for that reason reward participation, because that's what I want from my players. I could still use BG's approach, just that everyone who participates gets 1 point, then the top finishers get bonuses. (If every tourney was maxed out and there was always a waiting list, then I might drop the participation points!)

I also think that in, for example, a 20 player tournament you shouldn't get more points for finishing 16th than 17th, cause there really isn't any difference. Heck, the one finishing 17th might be on another table and actually lasted longer (in terms of played hands) then the one finishing 16th! I'd probably have small point jumps at table breaks instead, if I wanted to differentiate between early and middle finishes.

My real point is that any point system that is used will add an artificial reward, so make sure you reward what you want to encourage. If you give a good chunk of points to consistent players who regularly make deep runs but rarely cash (and then usually min cash), then your model means that they are better players than those who often bust early but occasionally win, even though the latter make more money.

tl;dr: Points are fun but add artificial rewards. Make sure to reward that which you encourage.

 

[Gobbs]

First, I totally agree awarding low finishers more points than lower finishers is not good. Somebody who finishes 9th did no better than the last place finisher at 16th. It’s a matter of strategy and style. I could let myself get blinded off and finish ninth (or better), but the person who finishes last may have an aggressive style that either gets beat early or earns tons of chips early. The person who finished 12th took a really bad beat. Did she/he do better than the two others? I give them nothing (although giving them all one point is not horrible).

Theoretically, if it’s all about money, let that be your point system - dollars earned. It wouldn’t be the worst idea, but I like giving points a bit more liberally than dollars. I think @BGinGA ’s 25 percent cash/33 percent point is about right for home games.

 

[Taghkanic]

Below is the current state of my table. We are nearing the end of a six-month cycle in which we were not using a points system, so I’m close to implementing this.

The formula — which will seem arcane, I’m sure — is, where:

(((t−p+1)^1.28) ÷ (t+1)) + 1​

Where:

t = total players
p = place​

I’ve cut off the chart where the points for the lowest place stays at 1.

There are no doubt some things about this which will rankle some. Why, for example, should someone placing 1st in an 18 person tourney get more than twice the points of someone placing 1st among 9 players?

I’ve done this because my sense, playing with my particular group, is that it is more than twice as difficult to win a two-table tournament than a one-table tourney.

In the latter, players essentially only have to eliminate their opponents once to ladder up meaningfully closer to the money. With more than 9 players, as tables get balanced, eliminated players are in effect replaced by a fresh player (who may be shortstacked, but more likely will be coming in with substantial chips), and you have to start all over again as if starting a “new” 9-handed game, but with uneven starting stacks. Multi-table tournaments also inherently last longer than single table ones, and require an additional set of skills beyond single-table tourneys. For example, in a 17-player game, there is going to be a period spent playing shorthanded but relatively deep in the middle stages, as you wait to combine—which is typically different than how a single-table game plays out, where you don’t get into shorthanded & shortstacked territory except quite late.

Not explaining that well, and I know some will disagree with my chart. I’d value some suggestions before putting it into practice. But this is the result of a lot of tinkering.

P.S.: The reality is that our game averages 14-15 players, seldom if ever fewer than 12, and never more than 18. So the lower and higher portions of the table are more notional than anything else. As such, the progressions of points are more tightly grouped in practice than it might appear from the 9-27 player table.

 

[Taghkanic]

First, I totally agree awarding low finishers more points than lower finishers is not good. Somebody who finishes 9th did no better than the last place finisher at 16th.

On this point we disagree... though I suspect such differences of opinion may have to do more with the particular player pools in different games.

The game I host is almost entirely regulars. No one ever just tries to fold for the first two hours to place somewhere in the middle. But more importantly, looking over nearly four years of results (and also having been the record-keeper for the game when it was at a previous venue), the fundamentally sound players almost never finish in the bottom 3rd.

By contrast, the weakest players frequently go out in that bottom third, often before the first break. (Weirdly, and happily, they are also the ones with the best attendance!)

The top performers are hardly all “tight” players, and the worst performers are not all LAGgy maniacs. In fact, the guys who bust early tend to be the passive, station-y types who make the most basic poker errors (playing way too many hands, limping too much, seldom applying pressure, not balancing their bluffing ranges, overvaluing middling made hands and chasing bad draws, being easy to read, failing to take position into account, just playing their own hands without regard to their opponents, etc.).

In short: In my player pool, finishing 9th generally does mean that person played a lot better than the one who busted 16th. Of course, anyone can run into a cooler early on. But the only time I ever went out last in this game was years ago when on the very first hand I flopped top set against an overpair, the turn was a brick, we got it in, and the other guy rivered a higher set. By contrast, there are two regs in my game who frequently are among the first 2-3 out.

If I were running a larger and more public game with a wider cast of characters, I might be more likely to agree. As it is, if my regular player pool is about 20, and we average 14-15, I can look at who showed up for any given game and pretty confidently predict 4 out of the 6 top finishers. That isn’t accidental.

 

[Gobbs]

I can look at who showed up for any given game and pretty confidently predict 4 out of the 6 top finishers. That isn’t accidental.

I think you put your finger on the issue with the last line. If you can reasonably predict which players will finish highest, there is obviously a very huge disparity in your players’ skill levels. While I think all groups have disparate skill levels, I believe most groups are competitive enough that the differences in skill level between players takes longer to become apparent than just one or two tournaments.

There is no way I can predict four out of the top six performers in our tournaments no matter who shows up. Yes, there are a couple players I believe are obviously better than most and a couple I believe are less-skilled than most. Overall, though, the tournaments are far too competitive, with too many players very capable of cashing no matter who plays, to make anything more than wild-ass guesses. I would say even one season is not enough to reasonably show the differences in skill levels between players. In most leagues, I would say, other than the extreme standout or two and the extreme “other end of the spectrum”, it takes at least 25 tournaments to begin seeing true evidence of skill level.

I say all that because, if there is a large disparity in skill level, of course the standings shake out and show the weaker players finish lower. I don't think that is indicative of most leagues. In most leagues, players play well and still finish poorly.

 

[TexRex]

Some stats about what I think is a recreational group (primarily for fun -- no one is making enough to make a real difference in their lives, and no one is losing enough to make a big difference). Please note that some of these come from 2 different tracking spreadsheets and won't appear to match up exactly. That problem will be remedied soon.

In 6 tournaments this year, we've had 31 different players make 73 appearances; 25 paid slots. That's higher than normal payouts since we've had 3 tournaments where we had just enough to pay one more slot. We normally average paying about 30%. We've played 7 tournaments, but the stats from one hasn't yet been integrated into overall stats. That tournament had 4 additional players, 5 additional payout slot, and 19 additional appearances.
14 of those 35 have cashed at least once.
10 of those 14 (28%) are net positive; 1 player is exactly even; 3 who have cashed are net losers; 21 have not cashed.
1 player has cashed 4x.
2 players have cashed 3x.
4 players have cashed 2x.
7 players have cashed 1x.

Of the 21 who have not cashed this year ...
2 have been our top player in a given year;
2 other are net long-term winners;
Those 4 appear to be having an off year, which of course could change any time.

Since 2013, we've played 86 tournaments. 19 people have won 2 or more (either outright or by chop); 14 of those players still play with us at least some.
Another 14 players have won more than one tournament and chopped for at least 1 more, and 9 of them still play with us at least some.

Any of those 23 are capable of winning a tournament at any time, so well over half of the field has won in the past enough to make me believe they have a real chance. All of those are capable of making the final table (which is 10, and not that impressive when averaging 13.5 players).

Long term stat -- about 1/3 in any given year are net winners; and over 50% cash at least once in a year. We are slightly below both averages this year.

Our top 7 are "in the points," meaning they get more points than just for showing up.
20 of 35 have gotten in the points 42 (just under 52%) times.
2 have been in the points 4x, and that's the most in the points appearances.
6 have been in the points 3x,
4 have been in the points 2x,
8 have been in the points 1x.
Note: Making it in the points becomes more impressive the more a player does it.

27 of 35 players have made 56 final table (FT) appearances. The FT is generally 10, but we've had some tournaments with as few as 8, and one per year that only 9 players make the FT.
2 players have made the FT 5x (one of them is the break even player);
4 players have made the FT 4x;
2 players have made the FT 3x;
7 players have made the FT 2x;
12 players have made the FT 1x.
Only 1 person who has come 4x made the FT all 4 times; no one else who has come at least 3x has made it every time.
Note: I think making the FT shows some survival skills, and based on this limited data, you can see generally the number of appearances gets harder at the top.

I've been keeping stats since 2013, and we've had a different player of the year every year -- no one has won that a second time, at least not yet.

Statistically, we clearly have a group of players who are generally better than others, but not a group so dominant others don't have a chance.

I think we have a fairly competitive group, but I can't really compare it to similar size groups, so maybe we aren't as competitive as I think. But I certainly couldn't guess accurately who would be in the points or in the money. Even the FT is not certain for anyone.

Comments welcome.

 

[Taghkanic]

If you can reasonably predict which players will finish highest, there is obviously a very huge disparity in your players’ skill levels.

In general, I’d say there is not a giant disparity in levels of understanding of the fundamentals of the game among “my” gang, except at the very top and very bottom of the roster. There is more a disparity in levels of discipline, patience, attentiveness, impulse control, gambly behavior, etc. It has to do more with temperament than acuity.

Except, again, for the really bad ones. Luckily, these are also the guys who never miss a session, bless their hearts.

But to clarify, my comment was about predicting top and bottom finishers *among those showing up*, not among the whole player pool of invitees.

The top players are not always the same four people, but they tend to come from the same pool of invitees. If we have an average of 14-15, the player pool of regs is more like 20-25 people. Of those, there are 7-8 who are more consistently successful tournament players.

If 5-6 of those show up in a given week, I am going to expect most of them to make the final table, plus a few middling or even bad players who are running unusually good that night.

I attribute this less to the better players having Tom Dwan-level hardwired math skills, and more to a combination of discipline, experience and observation of the other players. In a longrunning game among regs, I find there is a lot more value to “playing the player” than just adhering strictly to opening range charts.

FWIW, some of the less successful tourney players in my game (though not the very worst ones) are good cash players. After all these years I’m still trying to figure that out... I’d say it has mostly to do with them just lacking patience for tourneys.

In most leagues, players play well and still finish poorly.

This happens, of course, in our game as well. Recently a guy who had finished 1st or 2nd in the previous five games went out quite early, running into a cooler. But again, I’m talking more about long-term trends.

As far as point systems and player pools, as a host I am not worried about losing either the best players (who have a profit motive to keep coming back) or the worst ones (who per above, and bizarrely, are the most regular attendees). It’s the guys in the middle who seem to experience the most frustration, as they are so often just missing the money and only occasionally breaking through to cash.

So for me as a host, a points system has to take into account not just mathematical purity, but what is going to keep people in the game. That includes recognizing attendance and middling finishes to encourage all types of players to keep trying, while also being equitable to the winners.

 

[Poker Zombie]

So for me as a host, a points system has to take into account not just mathematical purity, but what is going to keep people in the game. That includes recognizing attendance and middling finishes to encourage all types of players to keep trying.

This is why I built a points system. Do I think 2x 3rd place finished makes you a superior player? No. Do i think it makes you a wealthier player? No. Does the 2x 3rd place finisher hang out longer than the win-or-bust player? Usually.

Everyone in our group knows Ben is our best player. It's not even close. But a spread out point system gives everyone a fighting chance for the title, and that is just a little more fun, IMO.

 

[Gobbs]

In general, I’d say there is not a giant disparity in levels of understanding of the fundamentals of the game among “my” gang, except at the very top and very bottom of the roster. There is more a disparity in levels of discipline, patience, attentiveness, impulse control, gambly behavior, etc. It has to do more with temperament than acuity.

And I would argue that all those things, especially discipline and impulse control, are as big of a part of a person’s skill level as mathematics and fundamentals. I can teach a 10-year-old the math and the fundamentals (and have). The discipline and patience is another story and typically comes with experience and maturity. (Now, the finer points of the game, the metagame, and details is another story all together. When I learn them, I’ll let you know. )

Anyway, I digress. I say this to say I still believe there is a skill disparity in the group which explains why you can predict top finishers from the list of participants. Thus, getting back to the original point that this is the reason, in your group, the person finishing 16 of 16 is often the worst player, the person finishing 9th is generally mediocre, and the people cashing are quite often near the best. In many leagues, the outcomes are not as predictable, thus 9th place probably didn’t play any better than 16th place and neither deserve points.

 

[Taghkanic]

And I would argue that all those things, especially discipline and impulse control, are as big of a part of a person’s skill level as mathematics and fundamentals. I can teach a 10-year-old the math and the fundamentals (and have). The discipline and patience is another story and typically comes with experience and maturity. (Now, the finer points of the game, the metagame, and details is another story all together. When I learn them, I’ll let you know. )

Anyway, I digress. I say this to say I still believe there is a skill disparity in the group which explains why you can predict top finishers from the list of participants. Thus, getting back to the original point that this is the reason, in your group, the person finishing 16 of 16 is often the worst player, the person finishing 9th is generally mediocre, and the people cashing are quite often near the best. In many leagues, the outcomes are not as predictable, thus 9th place probably didn’t play any better than 16th place and neither deserve points.

You’ve kind of exaggerated my point here. I’m hardly arguing that our results are 100% predictable. No one would play if that were the case, except winners and addicts, plus maybe those willing to spend $120 to hang out.

But after tracking a history of several hundred sessions with a steady pool of regulars, yes, I would feel comfortable predicting with confidence the *likely* outcomes once I see which 14-15 guys show up on a given night. Give me 6 picks, and I should get about four right.

I might be off on a few picks due to extreme variance, and anyone can finish in any spot. The guy who has won exactly once in the whole history of my game—well, one night he did win. That says very little.

Anyway, really we are talking here about league point systems which run across long (6-12 month) cycles. Once one smooths out variance over that length of time (assuming 2-4 games per month) it should be easy for most hosts to bet confidently on who will make up the top 30% or so. A graph of everyone’s results long term is never going to be a flat line, even though no one will get better cards or flops long term.

If that were not the case, we might as well play Chutes and Ladders.

 

[Poker Zombie]

Another good reasoning for not paying early finishers points in a MTT is that players are facing completely different opponents. You may be sitting at a table with the next Stu Unger, and get eliminated far earlier than you would if you were at the other table with just average players. Is the player at the other table better and deserving of more points, just because he is at a softer table?

 

[Taghkanic]

Another good reasoning for not paying early finishers points in a MTT is that players are facing completely different opponents. You may be sitting at a table with the next Stu Unger, and get eliminated far earlier than you would if you were at the other table with just average players.

Presumably over the course of an entire League season, table draws even out.

If you wind up in any games with the next Stu zinger, my advice would be to stake him, but play in a different game...

 

[Taghkanic]

Presumably over the course of an entire League season, table draws even out.

There are nights when I see Player X has drawn a seat with Players A, B and C, and I know they are going to dump their stacks to him within 6 levels... And when we combine he’ll have a giant stack to run the rest of us over. (That might actually change how I play at my table.) But sometimes I’m the guy who gets the lucky draw, dodging the best players and sitting to the left of the donks.

If you wind up in any games with the next Stu Unger, my advice would be to stake him, but play in a different game...

 

[Poker Zombie]

In theory they could even out, provided the number of games in the season are equally divisible by the number of tables. That is not always the case, so it is more equitable to just award players that never played together the same number of participatory points.

 

[Taghkanic]

... so it is more equitable to just award players that never played together the same number of participatory points.

IDGI... How does this work?

 

[Poker Zombie]

IDGI... How does this work?

To make math easy...
You play 10 events over the season. 3 table MTTs. 100% participation from each player. Someone will play at Stu's table more than someone else. This means that player (or those players), while they may be excellent, will probably score fewer points, simply because they faced a tougher opponent.

If you only pay our the top half of the final table, everybody will have to face Stu. Yes, some had to face him longer, and they may not have even made the final table, but at least some of the variance of seating is mitigated by not paying 30th place less than 15th place. They both earn the same points for participation (which may be zero), because there really isn't any way to really claim that 15th was so much better.

 

[Taghkanic]

To make math easy...
You play 10 events over the season. 3 table MTTs. 100% participation from each player.

And if there is not 100% participation by each player?

 

[BGinGA]

100% of zero is zero. So is 50% of zero.

 

[DoubleEagle]

100% of zero is zero. So is 50% of zero.

And naught from naught is naught!

 

[Poker Zombie]

And if there is not 100% participation by each player?

Still likely that someone will be placed at Stus table more times than someone else. Even the rules of variance would indicate that someone will get stuck at his table more than others. Perhaps if you played 162 games, the variance would even out. In reality, most poker home games dont see more than 1-2 games a month. A point system that pays differing points to low level finishers is going to basically be awarding points for being seated at the right table.

 

[Taghkanic]

Still likely that someone will be placed at Stus table more times than someone else. Even the rules of variance would indicate that someone will get stuck at his table more than others.

So, what about negative points for being seated with the worst players more often than others? ;^}

 

[Poker Zombie]

So, what about negative points for being seated with the worst players more often than others? ;^}

That's called "college football BCS".

 

[Taghkanic]

Seating variance just seems to me a part of poker that players have to deal with, like being card dead for a couple hours, or running into set over set... I wouldn’t want to be bothered to do 30+ computations on who never sat with whom, just to award a marginal number of points (though possibly I could automate that in a spreadsheet).

As is often the case, if the goal of a series is to send the top player to the WSOP, whoever goes better be prepared to potentially get a bad (or good) seat draw. You might start out at a table of full of recreational rich businessmen just ticking off an item from their bucket list... or you might get Antonius, Galfond and Selbst. That’s just poker.

 

[Poker Zombie]

Which is more important: sending the best player, or sending the luckiest player?

Also sending the guy that finished 17th out of 18 because the other guy missed an event also feels wrong. If I were to send a player to the WSOP I would consult @BGinGA , right off the bat.

 

[Gobbs]

If I were to send a player to the WSOP I would consult @BGinGA , right off the bat.

Agree...and I wouldn’t even think twice about it.

 

[andy699669]

Another reasin I dislike most commonly-used formulas (including both shown above) is that they do not reward combined finishes properly. A player with one 1st and one 3rd should reflect a higher score than a player with two 2nd place finishes, and a player with two 3rd place finishes should never have a higher point total tha a player with a win and a 4th place or 5th place finish. And two 4th place finishes shouldn't be equal to a lone win, either.

Simplify:
(1st + 3rd) > (2nd + 2nd)
(1st + 4th / 5th) > (3rd + 3rd)
(4th + 4th) != (1st)

Example:
Points in Order of Placing (8UP): -10-8-6-5-4-3-2-1

Results using above scenarios:
Amber (1st & 3rd) 10 + 6 = 16
Boris (2nd & 2nd) 8 + 8 = 16
Cress (3rd & 3rd) 6 + 6 = 12
Daisy (1st & 4th) 10 + 5 = 15
Ethan (1st & 5th) 10 + 4 = 14
Fritz (4th & 4th) 5 + 5 = 10
Garth (1st) = 10

This proves the above flaws of the commonly used formulas, with Amber & Boris plus Fritz and Garths scores.

I typically use the 10-6-3-1 ratio of awarded points applied to the average number of players (for example = 12, in a two-table 18-player max league where the qualifying field size can range from 6 to 18 players) and adjust the awarded points up or down based on the actual field size. Cash is usually paid to the top 25% of a given field size, and points are typically awarded to the top 33% of the field size (and always less than half).

Fields larger than 12 increase the points by 5% per player; smaller fields decrease the points by 5% per player. For example, the awarded points for a 16-player field would be 12.0 - 7.2 - 3.6 - 1.2 - 0.3 and the awarded points for an 8-player field would be 8.0 - 4.8 - 2.4 , with no points for 4th in an 8-player field (always pay less than half).

Note that the point drops are always 60%, 50%, 33%, 25%, and 20% of the next higher awarded point total. For example, for a 10-6-3-1 payout, 6 is 60% of 10, 3 is 50% of 6, and 1 is 33% of 3.

For an 18-player field, points would 13.0, 7.8, 3.9, 1.3, 0.3, 0.1 (paying points to 1/3rd of the field, rounded to tenths where needed) which follows those same percentages (7.8 is 60% of 13, etc.).

If you don't like working with decimals (or just like really big numbers), multiply everything by 10. Personally, I like point races that are decided by a fraction of a point.

Hmm... That 1-3-6-10 is a pyramidal sequence, so for more players, what's the ratio using the 1-3-6-10 as the base? Is it 1-3-6-10-15-21-blablabla?

 

[BGinGA]

Hmm... That 1-3-6-10 is a pyramidal sequence, so for more players, what's the ratio using the 1-3-6-10 as the base? Is it 1-3-6-10-15-21-blablabla?

Yes, or it can be scaled to a more similar top-end number for multiple events with varying participation.

For example, if awarding points only to the top 33% (rounded) of a given field size, adding an additional 10% to the top finisher points for each extra player in the field, and using the average field size as the baseline for 10-6-3-1, then:
12 players (baseline), 4 places: 10-6-3-1
8 players, 3 places: 6.6-3.3-1.1
16 places, 5 places: 14.6-9.8-5.9-2.9-1.0
etcetera, vs a more simplistic 10-6-3-1, 6-3-1, and 15-10-6-3-1 application.

Most of our point structures use a +/- 7% multiplier for field size adjustments, which narrows the overall range of awarded points (16p-1st=13.11, 8p-1st=7.48, etc.).

Fwiw, we award points calculated to two decimal places (14.64, 0.97, etc. for example), pretty much eliminating any troublesome ties in the points standings.

 

[Taghkanic]

What’s the basis for saying that a 1st + 3rd is better than two 2nd place finishes?

I’m not necessarily disagreeing, just wondering what the thinking is. I’d tend to think of it in relation to ICM models. If someone finishes 1st, sometimes that is a decisive win; sometimes it is kind of fluky.

The other day I won a tournament where two of us had about 25% of the chips, and a third player had about 50% of them. The blinds had run up pretty high, so even the leader had maybe 25BB left. We were pretty much all in shove/fold territory.

I survived a preflop flip (AQs vs 88) against the other smaller stack... Now I just barely covered the remaining villian. Only a few hands later, we got it in pre AKo vs TT, and I won the tourney.

Did I really play significantly better than the other two? Not really. In both of the decisive hands, all the players made what I’d consider correct plays. I was dealt a couple premium hands in one orbit, against opponents who also had shoving hands. Really, any of us could have won.

Had it gotten down to three players where I had 70% of the chips, and the other two had tiny stacks, that might have seemed much more decisive—like I’d outplayed the room for much of the night. But it could still only reflect a lot of rungood, and not mean much. Even the worst player in any of our home games goes deep once in a long while.

Obviously, things settle back toward a mean result over time, with skill emerging eventually over normal variance. How best to reflect that in point systems? On average, in my game I’d guesstimate that 1st is more like coming in 1.5th, 2nd is something more like 1.8th, and 3rd more like 2.7th for a normalized field size between 9-18 players.

 

[karsus]

I made the decision to only reward the players towards the top with points, expanding the payout to be slightly more then ITM, but not by much. IMO, the difference between 10th place and 14th place is really not much and can easily be impacted by factors such as how many hands were played and the random strength of the table. I did want to reward the 1st + 3rd more than just two 2nds but again, only slightly. Below is the system I settled on (although it is untested).

1 point for attending
1 point bounty on overall point leader and 1 point bounty on prior tournament winner

For 10 or less players top 4 get points (6/4/3/2)
For 11-20 players top 6 get points (8/6/5/4/3/2)
for 21-30 players top 8 get points (10/8/7/6/5/4/3/2)

So in a 10 player tourney, a 1st + 3rd finish would get 11 points (7+4) vs the person who came in 2nd twice gets 10 points (5+5)

It still rewards the winner more than others, but doesn't make it overwhelming so

 

[andy699669]

Yes, or it can be scaled to a more similar top-end number for multiple events with varying participation.

For example, if awarding points only to the top 33% (rounded) of a given field size, adding an additional 10% to the top finisher points for each extra player in the field, and using the average field size as the baseline for 10-6-3-1, then:
12 players (baseline), 4 places: 10-6-3-1
8 players, 3 places: 6.6-3.3-1.1
16 places, 5 places: 14.6-9.8-5.9-2.9-1.0
etcetera, vs a more simplistic 10-6-3-1, 6-3-1, and 15-10-6-3-1 application.

Most of our point structures use a +/- 7% multiplier for field size adjustments, which narrows the overall range of awarded points (16p-1st=13.11, 8p-1st=7.48, etc.).

Fwiw, we award points calculated to two decimal places (14.64, 0.97, etc. for example), pretty much eliminating any troublesome ties in the points standings.

10UP Scores:
1st - ? (300)
2nd - ? (200)
3rd - ? (125)
4th - ? (75)
5th - 30
6th - 25
7th - 20
8th - 15
9th - 10
10th - 5
How many points do you suggest for the question marks? Numbers on parentheses are numbers I'm currently thinking at the moment.