Psychobabble's picture
By: Psychobabble, PB
Feb 26 2015 1:00pm
Login or register to post comments

There is a major problem in the payout of tournaments in magic online. Everyone who plays constructed knows this, and disquiet has been brewing for some time but we are now at a point where we have enough data to say that the problem isn't temporary or restricted to one set, a very clear pattern has emerged over the past two years and looks like it's here to stay. The fundamental problem is that the entry fee for constructed tournaments (unlike limited) is a fixed quantity of tickets, but the payout is in packs of highly variable value. For a very long time this system sort of spluttered along and worked ok because the secondary market value of packs tended to revert to a relatively stable and decent value. That is no longer the case, and something must be done about it for the future health of constructed MTGO tournaments. This article analyses the data that is available to show the nature and extent of the problem, speculates as to some of the causes and then suggest some solutions. This is a call for action. I hope to present enough evidence and information to galvanize the community around something that they know is a problem but might not have a big enough picture to really see what's been going on. I hope it achieves its purpose.

Part 1: The Problem

When an organisation runs a competitive tournament with entry fees and prizes, it creates an incentive structure for participants in that tournament. A deal is struck - you pay X entry fee you have the potential to get Y prizes in return. The framework of this structure for constructed tournaments has been settled on MTGO for many years. You have two types of constructed tournaments. On-demand queues fire as often as there are participants, and pay out 50% of their entry fees in pack "value" given the retail price of the packs (USD$4 is roughly equivalent to 4tix). All on-demand queues have this payout ratio, including 2-player, 8-player and the special "gold" queues. This structure means that on "retail value", you break even on both 2- and 8-player queues with a win rate of 50% which seems generally fair (all EV calculations in this article use the excellent MagicEV.

Daily events are a bit different. Because they don't fire at will, it's possible for WoTC to pay out more in the retail value of packs than they take in tickets without implementing a de-facto retail price decrease of packs (if they ever paid out more than 4-tix of prizes for an on-demand queue, people could camp both sides of the queue and forfeit matches to get better-than-retail prize payouts). The math is a bit complicated, and varies with tournament size, but on the existing daily event structure it works out that daily events pay out approximately 8.75tix in retail pack value for every 6tix entry fee they take in. Players can "break even" on this prize structure with a win rate of around 45%.

All of the above discussion was on the basis of retail value of packs. Anyone who has played on MTGO for any period of time knows that that's the wrong way of looking at it in the real world because the secondary market value of packs that are being sold in the store is almost always a fair bit lower than the retail price. That creates a potential problem, because you now introduce the potential for the prize payout of tournaments run by WoTC to change based on secondary market factors that aren't under their direct control. By happy coincidence, this has generally not been a major long term issue over the history of MTGO. There has been a general price trend with most packs that are getting paid out as prizes. The trend has been most pronounced for the first set of a block, but has held generally true for other packs for sets that are in standard as well. The way the secondary value of prize packs has gone historically is that it start out in the high 3tix (3.8 or 3.9) and then slowly declines for the whole time that pack is being paid out as prizes to a floor of approximately 2.7-2.8. Once that pack stops being paid out as a prize, the value of the pack slowly creeps back up for the whole period that the relevant set is in standard, up to a maximum of around 4tix. Post-rotation of the relevant set, the behaviour of packs gets less consistent but at least for the time that the set is in standard, this has been the usual historical pattern.

There's not much hard data to support this trend, unfortunately, because we don't have a reliable MTGO secondary market data set that goes back further than July 2012. I'll highlight that data, from MTGGoldfish, though. First of all, here's the two fall sets that fall within the data set (I stop at RTR for reasons I'll explain later):

You can only see the full cycle for RTR, but it was the same pattern with INN and you can also see that Scars of Mirrodin packs were 4tix at the end of that set's time in standard prior to the release of RTR. This has been the typical price path for fall sets for many years on MTGO. If you listen to Brian Wong's explanation of "going infinite" on MTGO he goes into this. He has been playing essentially since MTGO began, and one of his key "secrets" to going infinite was to amass a huge pool of pack prizes while the set was being paid out as prizes and then selling them the following year - essentially rolling over his pack inventory in a long term spec based on a predictable movement in pack prices driven by fundamental secondary market supply and demand for the packs. The fall set is important because it's (usually) the pack that is paid out in constructed prizes for the longest amount of time, but the general pattern of "value decline while being paid out, followed by a slow recovery of value while the set is in the standard rotation" is also observed with other packs prior to and just after RTR but things start to change after that, most notably with DGM:

As you can see, AVR declined to 3.2 before rising back to around 4 until the set rotated. M13 dipped to 2.8 before rising back up to 3.7 and then hovering a little lower at around 3.5. Even GTC (the first set after RTR) maintained a fairly health price after its initial dip. However, the wheels started to fall off pack prices after that, with Dragon's Maze having a very rapid decline down to 2 and then trending down further over time:

Things went completely to hell, though, with Theros:

The price path for THS up until the release of BNG was entirely in line with the historical norm for the fall set, settling at around 3.2 when it stopped being the main pack paid out in prizes. The usual bump after the release of the next set, though, never materialised. THS packs just kept on falling. And falling. And falling. Until they are now worth a paltry 1.2tix - less than most out of print sets. This is despite it being a currently redeemable, in-rotation, standard set. It's not a low value set, either, being worth 20-tix more than Khans and having expensive chase cards like Stormbreath Dragon, Elspeth, Sun's Champion and Ashiok, Nightmare Weaver. BNG similarly had a big price fall that it never recovered from, and while the next small set (JOU) maintained a decent price, that's really looking like an aberration when we look at the price path of the current fall set, Khans of Tarkir:

Khans has fallen in value much further and faster than fall sets typically have, and it shows little sign of rebounding almost a month after FRF's release. While there's some chance of it bucking this recent trend, I'm less than optimistic.

What does all this mean? When packs decline in value - and there does seem to be a structural decline in the secondary market value of packs in recent years - tournaments become less attractive. This is not good for players, and it's not even good for WoTC because it's not like they're making any extra money from the fact that there's now a 2tix+ gap between retail and secondary market prices instead of a ~1tix or less gap. Because of this, less players are entering tournaments than otherwise would which means less players are having fun in these competitive events and less players are paying entry fees to WoTC. ie. Unhappy customers and less revenue. This is a classic lose-lose for all parties concerned. The situation is so bad right now that in the most popular tournament on MTGO, 2-player queues, with a THS prize pack value of 2.6 a player needs to have a seventy five percent win rate in order to break even. That's atrocious, and unlike in the past there is no expectation that you can hold onto that pack and have it be worth more down the track. In the recent past when prize packs have been worth as low as 2.2 (or less!) the required break-even win rate gets into the utterly absurd 90%+ levels. Quite frankly, it's a wonder (and a sign of how addictive this game is!) that anyone plays them at these levels.

The key message to take from the above data is that this isn't a short term problem by now. It has now been almost two years since the issue first started coming to a head, and it has affected the all-important first set of both of the last two blocks. This is a real problem that must be addressed, it won't go away on its own.

Part 2: The Cause

The cause of this recent trend is very difficult to ascertain without the type of data that only WoTC has access too. Many people point to increase of the redemption shipping fee from $5 to $25 after RTR as the cause, and that could certainly have had an effect. This move effectively reduced the overall value of MTGO cards - which is ultimately driven by the redemption value - by a significant proportion (perhaps 20-25% depending on what the set value would otherwise have been) and is likely to have had a flow-on effect to pack value. I think this tells only part of the story, though, if at all. In the post fee-increase world, a number of sets have still maintained a reasonable value. Specifically GTC, M14, JOU and M15 all maintained pretty good pack values over most of their time in standard. Another related explanation is that WoTC might just have been releasing more "bad" sets from a value perspective. This explanation has some merit given that the first recent set to suffer a major value decline was DGM, which had almost no value outside of Voice of Resurgence. I don't think that this can account for the decline in the value of the two big, value-packed, fall sets that have been released since then though.

Another superficially attractive explanation is pack payout ratios. The main source of secondary market pack demand is limited events. Outside of sets which are triple-drafted (or which make up all of the packs in a 6-pack sealed), there is an imbalance between the number of packs of each set demanded by drafters. It's virtually impossible for WoTC to match up the pack prize ratio throughout all of their constructed and limited events with the ratio used by limited players, which has meant that even older sets like DKA have suffered major value falls when the prize payout ratio was weighted too much in one pack's favour. WoTC often adjusts payout ratios to try to address this, most notably changing on almost a weekly basis for a time there with THS and BNG, but at the end of the day this issue might lead to temporary issues but can't be responsible for the decrease in the value of the fall set (which is triple-drafted) and can't account for the generally sustained decline in the pack value of the other sets. Some people also try to say that V4 has something to do with this, because that's a convenient whipping boy for all of MTGO's ills. This obviously can't be the explanation because the timing is all wrong (the issue began well before the V4 switchover) and because a reduction in playing population would have an almost symmetrical effect on both supply and demand and so can't really affect pack prices one way or another.

Believe it or not, I actually think that the most likely significant cause is that MTGO has been getting better. Ever since the release of MMA - which, crucially, was soon after DGM was released - WoTC has been doing much more to create a more varied online drafting experience. This has come from sets specifically created for drafting - MMA, VMA and the just-announced Tempest Remastered - as well as a much greater number of special drafting events. This includes a more frequent, longer lasting and better cube experience, holiday events and large blocks of flashback drafts. All of these moves have been hugely popular with the hardcore drafting population on MTGO. I believe that the type of people most attracted to these new draft experience are predominately the type of die-hards that used to grind out the 50th, 100th, 200th draft of whatever the most recent draft set was (unless it was as bad as AVR - but even then, that was a super high value set that attracted draft value junkies!). As noted above, limited events and specifically on-demand drafts are by far the main source of secondary market demand for packs on MTGO. This move to create a greater variety of events that cater to the high-repetition drafters has quite possibly lead to a substantial reduction in overall demand for the packs that are being paid out as prizes in constructed tournaments. I'm not for a second suggesting that this is a bad thing, but I do honestly believe that this is the most likely cause of the recent phenomenon of massive, sustained, tanking in the price of packs (especially for the important fall set). If this is indeed the case, it's yet another reason for WoTC to fix the system - doing something to make one part of their customer base happier seems to have had a major negative effect on another part of their customer base (with no real offsetting revenue benefit from the constructed crowd), and that doesn't make good business sense.

Part 3: The Solution

This is not a problem that's easy to fix, and I don't think it can be done without a relatively major overhaul in either the way prizes are paid out or in some of the MTGO system. I'll outline a few possible solutions and my comments on them below.

  • Option 1: Tweak pack payouts.
    The way WoTC has typically tried to address this problem before is by changing pack payout ratios. As I mentioned above, they did this multiple times with THS/BNG packs, but ultimately to no effect as the value of both tanked. Doing this is an extremely band-aid, short-term solution and doesn't do anything to address what has become long-term, underlying issue.
  • Option 2: Major overhaul to pack payout system.
    One option that still keeps within the current framework of prizes would be to change pack payouts in constructed events from a fixed type of pack to any type of regular set pack that currently being sold in the store. The basic idea is that rather than being paid out in the most recent sets only, players would be able to choose which of all of the current standard-set packs they get paid out in. That way, people will dynamically select the pack which they most want, either because it has the highest secondary market value or because they want to draft (or even open!) it. This wouldn't be easy to implement UI-wise, perhaps it would need a "store credit" type system, but it's the option that best fits within the current concept of pack payouts while still having a reasonable chance of success. I fear that as with option 1 though, this might prove to be a band-aid solution too and you'd just end up tanking the value of all available packs, but in an even more confusing and frustrating way. Still, it's worth a thought.
  • Option 3: Pay out prizes in tix.
    This directly and completely addresses the issue of "fixed entry, variable prizes". If events paid out in tix instead of packs, you'd be in a situation more like real world GPs where the prize payout is clearly defined and certain. I can't imagine that this raises any gaming/gambling type issues, this is still a digital game giving out digital products that have no official real world value. As with Diablo gold or WoW items, the fact that there is a secondary market outside of the game that puts a $ value on these products doesn't change the fact that they are digital game items with no official value. There's a couple of disadvantages of this option. First of all, it would have a huge effect on the secondary market value of packs because the main source of supply would be gone - most likely it would drive all of them up to or very close to full retail value. That would have a relatively major effect on drafters. Speaking of draft, it would also mean that you can't use your constructed winnings to directly free roll your drafts. It also does in a real sense take away some of the "prize" feeling of the current system and would make tournaments seem more like grinding out currency. Still, this is an option worth thinking about because it's the only one that would completely eliminate the problem.
  • Option 4: Pack redemption

    Now, I'm not talking about actual redemption in terms of swapping digital packs for real world packs. That would be both hugely unwieldy and open up a massive can of worms in the paper pack economy. What I mean by this is that the MTGO store should allow players to sell their (current) packs back to it for a defined amount of tix. A simple value, that tracks closely with the normal historical price of current packs, would be 3tix. So you could just go into the store, click "sell packs" and get 3tix back every time. You could select other partial tix values by making players sell groups of packs for a set amount. Eg. 10 packs for 28tix is a per-pack rate of 2.8. The effect of this would be to put a price floor on the secondary market value of these packs. There would be essentially no effect while the secondary market value was higher than the store redemption value. But it would (assuming people are rational) stop the secondary market price falling below the store level, or secondary market trades would just dry up at that point.

I honestly think that option 4 is the way to go. It is very likely to have the desired effect and still preserves the fundamental nature of the existing prize system. It will have some impact on drafters, who are currently enjoying very cheap entry costs. But these costs are extremely anomalous by historical standards, and drafters have survived under the old system. Plus the increased cost of entry is somewhat offset by an increased value of limited prize payouts (particularly relevant with events like sealed dailies). I know that this would be a major business decision for WoTC to take, and would obviously have to be looked at very carefully, but the status quo is not an option. There is something fundamentally broken with the current constructed prize payout system and something needs to be done. I know the issue of prize value is on WoTC's radar (eg. here). I hope WoTC will take this information into consideration very carefully and do something to fix what appears to be a permanent, long term, problem once and for all.


I agree man the state of by MrWishyWashy at Thu, 02/26/2015 - 15:53
MrWishyWashy's picture

I agree man the state of MTGO's prize support is strange. Especially for Eternal formats to where there is little to no incentive to spend lots of money on more expensive decks if your main goal is to gain as much EV as possible.

Also I had a question about the writing on this site. I just started submitting articles and cant upload decklists. How do you do that? Also where is the contact info for the editor or the people who run this site? I cant find that anywhere.

Thx for the comment. by Psychobabble at Thu, 02/26/2015 - 16:21
Psychobabble's picture

Thx for the comment. and are what you are looking for.

Greatly appreciated. I should by MrWishyWashy at Thu, 02/26/2015 - 16:51
MrWishyWashy's picture

Greatly appreciated. I should remind that person to put that in the write for mtgovideo since its nowhere to be found on this site.

Once I read that paying in by TugaChampion at Thu, 02/26/2015 - 23:32
TugaChampion's picture

Once I read that paying in tix would actually have issues with gambling laws but I have no idea if that's true or not.

One way to get around that would be to make constructed tournaments that cost 6 tix accept 6 tix or 2 tix + 1 pack that is currently sold in the store.

Making Release events accept 6 packs + tix would also help delay the decrease (it used to be like that in Zendikar and before).

And there is another way to help with this. Right now DEs pay 6 packs to 3-1s and 11 packs to 4-0s. Replace one of those packs (and one only) with something that could would be worth 4 tix. A special thicket or something like that, that could be used to enter tournaments, replacing 4 tix of the cost. By giving out just one, it makes sure no one can ever get more of those than they can spend and it decreases the number of packs being paid out.

The old 4 boosters sealed was also a very good booster dumper and it going away also contributed to this problem.

I agree with you that the main cause is probably all the Cube drafts plus MMA, VMA and other old draft sets but obviously those need to stay. I don't think it's the only cause, though. I have no way of knowing but maybe V4 is also affecting one side more than the other. We also had some time without Daily Events after the Kibler/MOCS incident which caused massive decreases in the overall value of cards and caused discomfort to many players. And we were without MOCS and PTQs for a long time. All that matters.

Before the Kibler/MOCS incident I used to be able to sell tix in my country for €0.90 each. This is because when buying in the store, we pay taxes making them much more expensive. People could probably find them a bit cheaper but they bought from me or another grinder because they knew us. After the incident the price of tix also crashed. People started buying for 0.80, 0.75 even going to 0.70 maybe 0.65. I didn't sell any for less than 0.75 but that was because I held on. Now I can easily sell for 0.80 so it's getting back up and it means people are becoming more confidant in MTGO again but it's still a long way from what it used to be.

Good point on 4-booster by Psychobabble at Fri, 02/27/2015 - 22:20
Psychobabble's picture

Good point on 4-booster sealed. My recollection is that they ditched them during GTC though? As with the redemption fee increase, I would have thought that would have impacted GTC prices if they were the major cause. it's a definite possibility though, and it's one of the "demand-side" causes that I think have the biggest impact.

One thing they did frequently by TugaChampion at Sat, 02/28/2015 - 06:51
TugaChampion's picture

One thing they did frequently in V2.5 was nix tix drafts of the latest sets. Packs would increase during that time because so many more people were grinding drafts. Too bad they haven't done those anymore. It would be good for both limited and constructed players.

The briefly got rid of 4 pack by Misterpid at Sat, 02/28/2015 - 12:55
Misterpid's picture

The briefly got rid of 4 pack sealed tourneys when GTC came out. After about 2 or 3 months they were brought back.

I haven't really paid by TugaChampion at Sat, 02/28/2015 - 14:11
TugaChampion's picture

I haven't really paid attention to that. Are they still exactly the same as before?

When the were brought back, by Misterpid at Sat, 02/28/2015 - 14:35
Misterpid's picture

When the were brought back, they added 2 tix to the entry fee and the winner got 1 more pack.

Yeah they're much worse now, by TugaChampion at Sat, 02/28/2015 - 14:44
TugaChampion's picture

Yeah they're much worse now, not only it's worse EV but it also consumes tix while paying one more pack. This makes them less played and a worse pack dumper.

It was possible to go by Misterpid at Sat, 02/28/2015 - 16:12
Misterpid's picture

It was possible to go "infinite" with the original cost/payout. I'm not a great player, but once I went 6 weeks of playing at least one M12 4 pack sealed every day without having to pay. That was the only time I ever went more than 2-3 events without having to pay anything. If I was capable of going 6 weeks free, better players were capable of that type of thing more frequently.

WotC felt the need to lower the potential for going "infinite" in the 4 pack sealed events - so they came up with a lame excuse to get rid of them. After the expected outcry and the fact that neither the 6 pack sealed nor the phantom sealed events were bringing in the revenue that 4 pack sealed generated, the 4 pack events came back with the new cost/payout. I'm sure that raising the price was what they had in mind all along, but the way they went about doing it was an insult to the intelligence of the people who played those events.

That's what they usually do. by TugaChampion at Sun, 03/01/2015 - 11:45
TugaChampion's picture

That's what they usually do. But even if some people went infinite in those, there were other players not going infinite and it was a good booster dumper that we don't have now.

Great article, perfectly sums by Mundisv at Fri, 02/27/2015 - 08:44
Mundisv's picture

Great article, perfectly sums up a huge problem for us, Constructed grinders. Funnily enough, just as you wrote this article, pack prices raised up to 4+ for FRF and 2,5 for KTK :).

Out of the solutions you proposed, the 4th one seems the best and TugaChampion's solution (reducing the number of actual packs given) sounds reasonable.

Since publication, KTK has by Psychobabble at Fri, 02/27/2015 - 22:19
Psychobabble's picture

Since publication, KTK has fallen down to 2.2.... we are now in the utterly ridiculous territory of needing 90% winrate to break even on 2-player queues.

Great article, on point and by Fragoel2 at Fri, 02/27/2015 - 09:04
Fragoel2's picture

Great article, on point and with some reasonable solutions.
You should tweet it to Lee Sharpe and Worth, see if they say something

Interesting article, I think by thewoof2 at Fri, 02/27/2015 - 11:20
thewoof2's picture

Interesting article, I think we also have to look at this from WOTC's perspective. The problem to them would be if the current pay-outs result in a lose of players revenue that is GREATER than the lose in revenue in any of the solutions. I wonder if we have any stats that show constructed events fire LESS often when the packs they pay are worth less. WOTC for sure has this so at least they can see if indeed the problem exists at all in mass. But then the question becomes are any of the options you laid out revenue favorable or at least neutral to WOTC.

I was intrigued when you mention "Players can "break even" on this prize structure with a win rate of around 45%." I think this is key, the question becomes what is the right % to attract players yet maintain a strong profit margin. We are assuming it should be 45%, and I like that as well (only have to win half the time is nice) but maybe we are working under the wrong assumption that WOTC wants it to be 45%.

Just to be clear, that 45% by Psychobabble at Fri, 02/27/2015 - 22:21
Psychobabble's picture

Just to be clear, that 45% comment was in the context of 4-tix "retail" pack value. if packs are worth 3tix, it takes about 50% winrate to "break even" on dailies.

I was thinking about how it by guitarweeps at Sat, 02/28/2015 - 15:23
guitarweeps's picture

I was thinking about how it would be nice to be able to select which packs you got for prize. I was thinking current block but with the new structure should probably be any in print set. Also the store credit or some partial credit would be cool and would provide a way for grinders who don't buy from the store to get supplemental products like commander and duel decks.

You mentioned paying out in by el_pato at Mon, 03/02/2015 - 00:42
el_pato's picture

You mentioned paying out in tix. Another solution is the reverse of this - pay in with boosters. Imagine a 2 man queue that pays two KTK and costs 1 KTK per player. That will ALWAYS be a fair deal. Or a Daily that offers the options of either 6 tix or 2 packs of any Standard legal boosters. This would not be difficult to implement since we already use boosters for some entry fees. And they would not have to calculate a fair option for every single event; just giving this option for dailies and two mans is enough to help eat packs and foster steady, rewarding competition without being overly generous.

Interesting idea, I like it by Psychobabble at Mon, 03/02/2015 - 01:54
Psychobabble's picture

Interesting idea, I like it too.