xger's picture
By: xger, Xger
Oct 05 2016 11:00am
0
Login or register to post comments
1963 views


Whether it's the upcoming Treasure Chests, the "cheaper" limited events, or the redemption changes, Wizards' recent upheaval will likely have wide ranging impact on MTGO and its economy. Quantifying that is more art than science given our lack of data and precise information. At first, Wizards gave us barely information, but a little less than a week later gave us more information. Time to analyze what we have! 

What did Wizards do?!?!

There are three primary changes that were announced on Thursday the 29th:

  • Treasure chests (link here)--This is where the Kaladesh Inventions, Commander 2016, and similar will enter the system (curated card list here)
  • Redemption availability (link here)--Shortly (yes, pun intended), the redemption period is being cut ~75% with a much harder deadline
  • Tournament changes (link here)--Draft is better, sealed is worse, constructed is ambiguous. The constructed changes are discussed in the treasure chest link, not the tournament one.

The response was swift--mostly negative with some willing to wait it out. Very little purely in support of the changes. So, Wizards let us know changes were coming.  That announcement came Tuesday the 4th: The changes aren’t really anything. The only topic was chests, no mention of tournament changes or redemption. We get to trade chests, and Wizards published the numbers for the current chests. Despite a lot of early conjecture that the likeliest split of Modern rare/mythic, curated, and play points was 1/3 to each, the numbers are actually far worse. 51% chance for a Modern rare/mythic, 30% for play points, and 19% chance for a curated card.

First up, Treasure Chests:

Wizards is introducing Treasure Chests, which are untradable objects (for now) that you can open up for "treasure." 

These chests are randomized prizes, but do not expect much value, at least consistently--they are lottery tickets. Each chest has 3 slots, which are these:

  • One of: "Curated" card, Modern Set Rare or Mythic, or different quantities of play points
  • Standard common or uncommon. 2 in 9 chests will have this slot replaced with another from the first slot
  • Standard common or uncommon. 1 in 239 chests will also replace this slot with another from the first slot

Commons will be 2.5 times as common as uncommons (meaning, for Kaladesh cards, any individual common will appear ~3 times for each time an individual uncommon appears due to the numbers of the set). With the Rare or Mythic rare, the appearance should be what one would expect from opening packs--an individual rare is opened twice for each mythic from each set. However, it is unclear whether the ratio means all rares to mythics, or just rares compared to mythics in sets with mythics. In other words, if it makes one giant pool, you have everything pre-Shards of Alara that will significantly shift the balance, meaning mythics will seem rarer.

Curated cards--curated cards are simply cards Wizards has chosen to be in this special pool. The "curated" list is then curated again for rarity. Wizards has given us the weighting for the curated list. So, a Chittering Rats is 40 times more common than a Black Lotus. We also know that a card might be duplicated in the list for different versions--Blood Crypt versus Blood Crypt for instance, or Tarmogoyf versus Tarmogoyf. As for frequency of versions, we only know the Masterpiece Series cards separately. We now know that you are twice as likely to get the expeditions Blood Crypt than you are the Return to Ravnica Blood Crypt. Oddly enough, you are equally likely to get Arid Mesa as Arid Mesa. Finally, curated cards are all non-foil, except for Eighth and Ninth Editions which are all foil. Importantly, this means the Kaladesh Invention Chrome Mox and the Zendikar Expedition Arid Mesa going to be non-foil.

Fun fact: you are 3.33 times are likely to get Crucible of Worlds than you are likely to get Crucible of Worlds. Some Masterpiece cards are apparently easier to get.

It's important to note that the lists provided have had a number of errors. Originally, the only Scalding Tarn on the list was the expedition version. I pointed that out to Lee Sharpe, and he quickly replied, saying it was an error that would be fixed. So, if you genuinely believe there is an error, let them know!

We were also given a play points breakdown. 1 in 2201 times you will get 1000 play points. 68% of the time you will get 10, 15, or 20 play points and 27% of the time you will get 30 or 40 play points. The average is 24.76 points.

Second, Redemption Availability:

Here, Wizards was probably the shadiest in how they announced and framed the change. It's effectively impossible this is a change that is good for players without being disingenuous. So, what happened? Here's the redemption schedule for Shadows over Innistrad and Eldritch Moon:

Product Redemption
Release Date
Redemption
Guarantee Date
Redemption
Shadows over Innistrad 5/18/2016 10/25/2017 4/25/2018
Eldritch Moon 8/31/2016 10/30/2017 4/25/2018

 

So, you could theoretically redeem Shadows over Innistrad for nearly two years. Elritch Moon for about 21 months. For those that don't know, the difference between the guarantee date and the cutoff date is that Wizards guarantees you will get a set if you submit the request by that date, whereas the cut-off date is supply dependent. If Wizards overprinted a set, you could redeem for another six months. Wizards' announcement of the redemption changes conspicuously omits the new table with Kaladesh and Aether Revolt, instead giving a stand-alone image that really hides the dramatic nature of the change:

As nice and neat as this graphic is, it hides the truth of the change. Here is the new table:

Product Redemption
Release Date
Redemption
Guarantee Date
Redemption
Shadows over Innistrad 5/18/2016 10/25/2017 4/25/2018
Eldritch Moon 8/31/2016 10/30/2017 4/25/2018
Kaladesh 11/9/2016 6/7/2017 6/7/2017
Aether Revolt 2/1/2017 6/7/2017 6/7/2017

 

Notice the shockingly short date range and how clear it is here, but not in the article announcing the change? Seems like a rather purposeful attempt to hide the ball... Here is a fuller image to show the redemption

The short is: Battle for Zendikar is redeemable for 5 months after Kaladesh is no longer redeemable, and Shadows over Innistrad is redeemable nearly a year after Kaladesh goes off. In fact, Shadows over Innistrad will be redeemable for a few months after Amonkhet.

Third, Tournament Changes:

There are changes to both Limited and Constructed. Let's first overview the Limited changes:

Wizards has reduced the cost to enter limited tournaments! Drafts are now 12 Tickets/120 Play Points (or 3 packs and still 2 tickets) and sealed is now 24 Tickets/240 Play Points. Oh, and they also changed the prize structure on sealed:

Wins Friendly League     Wins Competitive League    
  Previous PP New PP Change   Previous PP New PP Change
 9 200 160  -40   5 420 360 -60
 8 200  160  -40   4 280  240  -40 
 7 200 160  -40   3 140  120  -20 
 6 200 160  -40         
 5 100 60  -40         
 4 50 -50         

So drafting gets mostly positive news, as it gets a price reduction but no prize reduction. Additionally, at pre-release, drafting gets an extra pack added to the prize with no extra cost! Sealed on the other hand gets a sizable deduction, but that deduction in prizes is ~35% of the deduction in entry fees for competitive and about ~78% of the entry for friendly, so overall it's a win (for competitive more so).

However, if you pay for draft with packs + tickets, you do not get the new discount! That could have deleterious effects on the price of a pack, though it will likely only be marginal at best, because grinders will very likely do the same as they did previously. In fact, the other changes might short the supply of packs, which would lead to an increase in value.

Draft leagues are also only going to be 6-2-2-2, the pack per win is going the way of the dodo. Apparently, at some point in the future we should get a "Competitive" draft league that is 8-4 single elimination and the 6-2-2-2 will be the "Friendly" league. No news on the time, but it probably shouldn't be too long given how quickly they've been updating leagues (and this is coming from someone who waited years for leagues to return...).

Also, only the current and previous blocks will be draftable going forward. The current block will be league, while the previous block will be only a 6-2-2-2 queue.

A last note: currently limited does not get the Treasure Chests. Because reasons. The only info we have gotten is that the chests are about putting cards into the hands of constructed players. Because only constructed tournament players need the cards. As I said, reasons.

Now, to the Constructed changes:

Let's get right to it. Here are the new costs and prizes. First, Friendly Constructed Leagues:

Friendly Constructed Leagues - Wins Old Prizes-Packs New Prizes-Packs Old Prizes-Play Points New Prizes-Play Points Treasure Chests Changes
 5 150  80  -2 packs, -70 PP, +8 Chests 
 4 120  80  -1 packs, -40 PP, +3 Chests 
 3 80  80  +1 Chests 
 2 40  40  None 
 1 20  20  None 
 0 10  10  None 

So, it will be much harder to 'go infinite' in friendly leagues, as you become entirely reliant on the Chests to make your value. Assuming Wizards equates 40 play points and a pack, the Chests would have to be between the equivalent of 18.75 and 26 2/3 play points.  Here is the Competitive Constructed Leagues table:

Competitive Constructed Leagues - Wins Old Prizes-Packs New Prizes-Packs Old Prizes-Play Points New Prizes-Play Points Treasure Chests Changes
 5 10 120  180  -4 packs+60 PP, +8 Chests  
 4 120  180  -3 packs+60 PP,+3 Chests 
 3 120  120   +1 Chests 

Here is the heart of the controversy (on the Chests). A 5-0 gives up ~2.5 packs for 8 chests, if you go with the likely Wizards assumption that 40 play points equals a pack. With that, a 4-1 gives up ~1.5 packs for 3 chests. A 5-0 needs to recover 12.5 play point equivalent per chest and a 4-1 needs to recover 20 play point equivalent per chest. Again, that assumes you accept the premise that 40 play points equals a pack (which is nearly never the case, but it would be a likely direction Wizards would point).

Last, there are some changes to the Vintage Daily Event. However, as that is such a niche topic (and this article will be quite lengthy), I am skipping coverage of that section.

Divining the Statistics of Chests

Based on my above, very rosy numbers, that 40 play points equals a pack. In reality, a single play point is not 10¢, but varies depending on the average value of busting a pack in a sealed event. Hearkening back to an old article of mine, we have this approximation for the value of a single play point:

 

These numbers are going to be ever so slightly off, as the cost of entry is down ~10%. This is also for the value of a play point once you have gone infinite. Essentially, if you are infinite, you can enter a sealed event just to bust the cards (then drop if you so desire), as this is really the only feasible way to convert play points into cards at any moment. So, the play point floor is about 3¢ (accounting for the entry difference and assuming a pack value of $1). Play points have this odd role--at certain points they are useless, then very valuable, and finally useless again (not enough, then enough to enter, then more than you can feasibly use). Because at certain times play points are worth 10¢ a piece, a middle ground seems reasonable. So, from here out, I will assume 6.5¢ per play point.

At 6.5¢ per play points, a competitive 5-0 needs to see 24 or so play point equivalents per chest. Here is the breakdown at 5-0:

Old Prizes: 10 packs, 120 points. $40/6.5¢ is ~615, so the total is 735 play point equivalent

New Prizes: 6 packs, 180 points. $24/6.5¢ is ~369, so a total 549 play point equivalent

735-549=186 play point equivalent difference

186/8 Chest=23.27 play point equivalent per Chest

For ease of calculations, and to include fudge room for lower value sets and the already approximated  play point value, I'll assume 25 play point equivalent is necessary for 5-0 competitive player to be at the same place. That means the average Chest needs to be worth $1.625 for the 5-0 to be in the same place, over the long term. [Ignore the 10 quantity in the next image...]

Now to the real issue: what is a Chest worth? First, let's collect the information we know:

  • There are only 4 uncommons or commons in Standard currently worth more than 5¢ and none worth more than 25¢. The average value is completely negligible, as the transaction costs will likely far exceed the value of those, so they should be excluded.
  • The 'good' slot occurs at least once per chest and up to 3 times. At 2 in 9 for a second 'good' and a 1 in 239 for a third, we have approximately 1.226 'good' slots per chests. With the standards cards being estimated at 0, this 1.226 'good' card makes up the entirety of value.
  • We have seen 10 chests opened. We also now have all the information necessary to fully approximate the EV.
  • The average play point payout is 24.76 (for those worried that the 1000 point payout skews this average, it is so rare, it doesn't skew it. The average without it is 24.31).
  • The chance to have a good slot occupied by a Modern rare/mythic is 51%
  • The chance to have a good slot occupied by a Curated card is 19%
  • The chance to have a good slot occupied by play points is 30%

First, Play Points have an easy to calculate value based on my above analyses. 6.5¢ * 24.76 = ~$1.61. So, if you get play points, the average value is $1.61.

Second, the Modern Rare/Mythic is a beast of its own. For this article and simplicity's sake, I will just use the number generated by a very helpful reddit user. The average value is $0.80. I may continue to monitor the value of a chest, and if I do, I will dig more into that and generate my own number.

Last is the "Curated" list. The list is 635 (including the gearhulks from Kaladesh, such as Combustible Gearhulk, which doesn't yet have an MTGO price). Here is the breakdown by rarity:

Curated Cards Mythic Rare Uncommon Common Bonus
Number 109 419 53 45 9
Total Value $679.07 $2827.80 $149.04 $80.62 $279.87
Average Value $6.23 $6.75 $2.81 $1.79 $31.10

The lowest value card is Luminarch Ascension at 3¢.Only 7 cards are sub 10¢. 132 cards are sub $1. Important for these calculations, 209 are less than $1.625. That means ~33% of the Curated card will not be enough to bring the Chest to a suitable value. I've seen some people mention how the prices are already crashing on these cards. For reference, I had run these calculations previously using yesterday's prices. From yesterday to today, the sum total dropped a whopping 2%. Easily could be related to the pre-release starting tomorrow. Also, it's possible my categorizations are slightly off due to the sometimes arbitrary rarity assignments by Wizards. Wasteland is rare, while Wasteland is uncommon for example.

For my numbers, I combined the Expedition/Invention frequencies and went with the lowest priced normal version. Since the new ones won't be foil, the price differential is unknown, but it might not be that much.

The average value overall, (as of around 12:00 PST on Tuesday10/4, going by MTGO sell price) is $6.33. If you are familiar with my Modern flashback series, you know that I try to hedge the lottery feel in order to give a more realistic view. The weighted average, accounting for Wizards "frequency ratio" is $4.97.

One last potential--removing lottery cards. Now that we have the 'frequency ratio' information' it'll be far easier to pick out lottery rares. It also means we get rid of cards like Timetwister, which is part of the Power 9 but only worth $7.63 also gets removed. If I just remove the ultra-rare (1 in 7649) cards, the average barely changes to $4.94. But, there I've only accounted for ~0.1% of the possibilities. So, I'll remove the cards with total 'Frequency Ratio' 1 and 6. That removes about 20% of the card pool. This, 'lottery-free' pool average is $4.34.

I'll also include the average value for each totaled frequency ratio. For example, the 1 range--which is the Power 9--has an average value of $31.10 and accounts for ~0.12% of the possible openings. Some are a bit wonky, such as the 26 frequency. This is the combined value of normal frequency and masterpiece frequency, where one is 20 and the other is 6.

Frequency Ratio Average Value Chance of Group
1 $31.10 0.12%
6 $7.35 19.83%
12 $6.09 39.82%
18 $6.89 3.29%
20 $0.75 4.18%
25 $2.13 19.92%
26 $7.78 2.38%
31 $0.30 0.40%
32 $3.01 2.09%
40 $0.74 6.79%
45 $5.95 1.18%

So, the values here don't look as bad when broken out. With a nearly 40% to hit a ~$6 card, that ratio will be the most often found. It's just the largest group of cards, holding 40% of the available cards. It's also good to note that the low value categories are such a small chance. For reference, the 18 group is the Expedition lands, the 26 group is a handful of the Kaladesh Inventions, 31 is just Tectonic Edge, and 45 is just AEther Vial and Lotus Petal.

As a reminder, here are the values used to calculate the total chest EV:

Curated Play Points Modern Rare+
19% of chests 30% of chests 51% of chests
$4.97/$4.34 $1.61 $0.80

The Estimated Value of a Chest:

Type of Average Value Average value of a 'good slot' Per Chest Value
No 1 or 6 frequency Chest $1.72 $1.83
Normal Chest $1.84 $1.96

So, after being given the numbers, we can see the EV of a chest really isn't that bad. The play points add $0.48, the Modern Rare/Mythic adds ~$0.41, and the curated cards give $0.94 (or $0.83 if you exclude ratios 1 and 6). Take those values and multiply by 1.226 (the number of expected 'good' slots) and you get the above Chest Value. Currently, if you accept my previous numbers, this means the EV is actually increased for the competitive 5-0 player. The competitive 4-1 player needed $2.60, a friendly 5-0 needed $2.44, and a friendly 4-1 needed $3.47. So, only the competitive 5-0 is up in EV, while others are down. A 3-2 in both gets a nice $2 boost though. I wouldn't be surprised to see chests hover between $1.50 and $2 once they are tradable, but that is just a guess.

Wrap-up:

Hopefully this gives some people a chance to relax and see where this takes us. On a plus note, Wizards claims to be reading all of the responses, positive and negative, to these changes. Change already occurred, as discussed above. At the least, I still strongly hope Wizards reconsiders the redemption changes or at least the severity. While a number of people focused on the constructed pay out and what that means, redemption has a far greater, and lesser known, impact on the MTGO economy. Some have theorized this is Wizards way of eventually stopping redemption all together so they can make MTGO significantly cheaper to bring in more people and get viewership similar to Hearthstone. Personally, I'm not sure I buy that, as MTGO (last we knew) still brings in significant revenue. Additionally, the problems with MTGO not being streamed to the same extent as Hearthstone or other games would require more than making the client free or nearly so. It would likely require a separation with paper, which would defeat the entire point of MTGO. For now, the sky is not falling.

Well, let me know what you think and if you want to chat about my numbers of methodology, please ask away!  Similarly, if you catch a mistake, please let me know. I've written this somewhat in haste, so it's entirely possible I missed something. 

xger

xger21 on MTGO

1 Comments

One addendum: I forgot to by xger at Wed, 10/05/2016 - 11:30
xger's picture

One addendum: I forgot to discuss that play points cannot appear in the second or third slot of a chest. Based on what Wizards gave us, it seems that there is about a 75% of a Modern rare/mythic and 25% of a curated card when the second and third slots are filled. So, the chest value is not 1.226 good slots, it is 1 good slot plus ((.75*.80)+(.25*.93))*(2/9) + ((.75*.80)+(.25*.93))*(1/239). This equals about 12 cents.