Fact check: Are pack weights changed secretly over time?

Share

A hot topic on Reddit recently has been pack weights. The term pack weight refers to the probability of packing a particular player. The topic became hot upon the community learning that multiple European countries[1][2] were introducing legal restrictions on in-game purchases just as EA was about to release this year’s TOTS cards. In this article, we try to shed some light on to pack weights and to kill a couple of the conspiracy theories surrounding the matter.

Lack of transparency = conspiracy theories

EA has stated that pack contents are random. That statement, however, doesn’t imply that you are equally likely to pull Messi and Lamela. It hopefully doesn’t come as a surprise to anyone that you are more likely to pull some players than others.

Up until FIFA 14, Dutch FIFA player jpflathead maintained the UltimateDB; a database containing information about how many new instances of a player where released to the market. This allowed him to perform crude assessments of the various players’ card weights relative to other players. If nothing else, this confirmed that some items are considerably more difficult to find than others.

Despite jpflathead’s good work, a number of questions in regards to pack weights remain unanswered:

  • What is the probability of packing Messi?
  • Do players maintain their pack weights when they get an upgrade or inform?
  • What determines pack weights?
  • To what extent does pack weights change across pack types?
  • Do all items with same same OVR have the same pack weight?
  • Does EA change pack weights secretly?
  • Are the same pack weights the same for everyone?

As is the case with the scripting and handicapping debacle, EA’s lack of transparency has nutured conspiracy theories. Although they in this case can be summarized into a common notion about EA changing the pack odds dynamically in order to make more money, there are a few variations over the theme:

“Maybe it have someting to do with current market price of player? Verthongen is cheaper then walker= bigger chance of packing verthongen.”
(– comment on Reddit)

Another player brought up the idea that EA secretly tampers with pack weights:

“The one thing I’ve been most curious about is if pack weights are affected at all by other people opening packs / earning cards as well.”
(– comment on Reddit)

And finally, there is this:

“Obviously new accounts get better packs, been like that for years”
(– comment on EA forums)

New data, new insight

A recent strawpoll conducted via Reddit did however shed additional light over pack weights. The strawpoll was released together with the EPL TOTS SBC, where players could exchange a TOTS player and an 83 rated squad for an untradable EPL TOTS player. As the poll is answered by an impressive 8,600 respondents (and counting), it allows for some statistically meaningful analysis to take place.

Do all items with same same OVR have the same pack weight?

It’s a fact that prices vary heavily for players with the same OVR rating. This could indicate that there are differences in supply, i.e. pack weight for players with similar ratings. For instance, Kylian Mbappe’s 83-rated card peaked at 230,000 coins during the autumn, while Naby Keita (83) had his peak at 5,600 coins.

On the other hand, most people realize that the OVR rating can be pretty misleading as it puts too much weight on stats with limited importance. Therefore, the price difference between cards like Keita and Mbappe could be demand-driven rather than supply-driven. Becausr of that, analyzing card prices alone wouldn’t provide a certain answer to whether the OVR influences the pack weight.

The recent strawpoll does however provide solid evidence here. Among the 8,600 participants, 4.0 % managed to get their hands on Fernandinho’s 93-rated TOTS card, while 5.9 % pulled Otamendi’s also 93-rated card. This difference is statistically significant (with 95 % confidence). An equally significant difference can be seen among the 92-rated cards, where the three most common (Vertonghen, Son and Matic) were found significantly more often than equally 92-rated Alli, Walker and Azpilicueta.

So, the results above leave no doubt that other factors than the card’s OVR influences the pack weight, as we clearly see players with similar ratings having different pack weights.

However, there is no doubt that lower rated items in general are found more often than higher rated items. The OVR is a strong predictor of a card’s frequency, but it’s not the only determining factor.

Are pack weights the same for special edition cards?

Another question which always has puzzled me is whether special edition cards (informs, TOTW, TOTS etc.) has the same pack weight as the original player item. In the dataset above, there are a few players who received a massiv upgrade when becoming TOTS. Most interesting is in this case is perhaps the 85-rated cards.

Mo Salah, whose TOTS-card is 98, was packed by 2,1 % (+/- 0.3 %) of the population, whereas Vertonghen and Fernandinho (upgraded to respectively 92 and 93) occurred in respectively 5,7 % and 4,0 % of the packs.

Equally interesting are 90-rated TOTS Young and TOTS Mané: Young is normally a 77-rated rare card, while Mané is 84 rated. Normally, I would expect Young (currently priced at 900 coins) to be pulled considerably more often than Mané, whose 84-rated standard edition currently sells for an impressive 14,500 coins. While TOTS Young appeared slightly more often than TOTS Mané in the strawpoll sample, the difference isn’t statistically significant.

Hence, the strawpoll strongly suggests that the probability of packing a player drops when he is available as a special edition.

Do EA change pack weights?

We know that EA has changed pack weights sometimes. During certain promotions, EA has announced that the chances of finding the best items were doubled, whatever that means. Whether they in fact did double the odds or not is of course unknown, though. What remains certain is that people have a bigger incentive to spend money on packs during promotions where the chances of finding usefuld items double.

But would EA have an interest in tampering with pack odds secretly? I have a hard time seeing this. It’s obvious that you can influence people’s purchasing behavior when informing them about pack odds having doubled. But doing so secretly is not going to influence spendings, because people don’t know it’s happening. You could of course speculate that EA would alter pack weights in order to ensure that certain items remain rare. But given that EA already controls pack weights, all items will be exactly as rare as EA wants them to be. Hence, there won’t be a need to fix this. And because of that, I would deem it extremely unlikely that EA would bother changing any pack weights secretly.

Are pack weights connected to the current market price or price ranges?

Another theory which came up in the recent pack weight debate is that current market prices determine pack weights or, even weirder, that price ranges determine pack weights. This is in my opinion an absurd theory.

It is beyond doubt that rare card also tend to be expensive due to low supply. But this fact clearly doesn’t lead us to conclude that these cards are rare due to their high prices. Rather, it would suggest that the cards are expensive because they are rare. In the same way, price ranges are set dependent on what people are willing to pay for a card. If they weren’t, cards would go extinct which is exactly what we saw in FIFA 15 after price ranges were introduced.

Hence, the true “chain of command” is that pack weights influence market prices, which in turn influences price ranges. It’s not the other way around as suggested by some of the people quoted earlier in the article.

All the things we still don’t know

While I feel that the recent straw poll did bring a bit more knowledge, thare are a load of things that we still don’t know:

  • The probability of packing Messi from a random packs bought with real money

We know the odds of packing all the EPL TOTS players from this SBC, but that doesn’t tell anything about what the odds of packing them are outside this SBC, and that obviously doesn’t tell us anything about the odds of packing all the other players in the game.

  • Exactly what factors determine the pack weight

The OVR is strongly corelated to pack weight in the straw poll above, but it doesn’t explain all the variation. We know that the pack type influences the chances of packing certain items as well, but apart from that, it’s likely that there is an individual modifier which could depend on the player’s position or the alternatives available in his league.

  • To what extent does pack weights change across pack types?

It’s obvious that gold items have a different pack weight inside bronze packs than bronze and silver items. Additionally, the game contains packs like the Electrum Players pack where each slot has a difference probability set for gold and silver items. But we don’t really know whether Messi’s pack weight is the same inside a Gold pack and an Ultimate players pack.

  • Are the same pack weights the same for everyone?

Do new accounts get better packs? Does your personal pack weights change depending on your mood, time of day in your location, age, whatever? We don’t know.

%d bloggers like this: