Reflecting on Pokémon GO Shiny Rates II: Raids, Eggs, and Quests

This article is the second in a three-part series where the Silph Research Group’s findings on shiny rates in Pokémon GO are consolidated. In the first part of this series, we examined shiny rates for boosted wild Pokémon encounters and identified three commonly-used boosted rates. Here are the highlights from the first article:

• The Community Day rate is used (you guessed it!) during the monthly Community Day event. Our 95% confidence interval (CI) for this rate is 24.4 to 25.1 encounters per shiny.
• The “Permaboost” rate is used for at least a dozen Pokémon that can be found in the wild and whose shiny rate appears to be permanently much higher than the usual rate. Our 95% CI for this rate is 56 to 65 encounters per shiny.
• The “Medium Event” rate is often used for Pokémon with significantly boosted spawns during events such as worldwide GO Fest and Safari Zone-related events. Our 95% CI for this rate is 126 to 141 encounters per shiny.

In this installment, we take a look at shiny rates for other encounter types: Raids, Eggs, and Field Research. Be forewarned — while there is strong evidence for many of the rates we’ll look at, other parts of this article may lead to more questions than answers!

---

Legendary Raids

Our Researchers have participated in more than 40,000 shiny-eligible Legendary Raids (not including special three-hour Raid events) and encountered more than 2,000 shinies, finding a shiny rate of

1 in 19.4

with a 95% CI of 18.6 to 20.2 encounters per shiny. This rate has been highly consistent across species.

Missing from this table are the Kanto birds and Johto beasts because data collection for those species has been entirely during their special three-hour Raid events. Which brings us to…

Three-Hour Raid Events

Niantic has held a series of 11 special three-hour Raid events, including one for each of the Legendary birds and beasts, as well as Gengar, Lapras, Lickitung, and the Kanto starter lines. Excluding the first of these events (the July 2018 Articuno Day, which appears to have had a separate, lower rate), our Researchers have found a shiny rate for these events of

1 in 10.5

with a 95% CI of 10.2 to 10.9 encounters per shiny. This rate has once again been highly consistent across special Raid events from Zapdos Day onwards.

On Articuno Day, in contrast, our Researchers found 194 shinies in 3,611 encounters, for a rate of 1 in 18.6 (95% CI of 16.3 to 21.5 encounters per shiny).

Other Raid and Egg Exclusives

A number of other Pokémon have been available in Pokémon GO exclusively (or nearly exclusively¹) as Raid bosses and by hatching Eggs. These include:

• Baby Pokémon (in Eggs)
• Certain Alolan forms, including Grimer, Meowth, Sandshrew, and Vulpix (in Eggs) plus, Marowak and Raichu (in Raids)
• Absol, Klink, Mawile, and Shinx (in both Raids and Eggs)

In this category we also include Kanto Regionals from their release event, during which they were available in Eggs worldwide. These rates do not apply to wild Kanto Regionals found after the event.

Our data for these Pokémon do not point definitively to a single, consistent shiny rate. For each of the above categories taken in aggregate, here is what our Researchers found (data broken down by species can be found in the appendix):

Combining the five categories together yields 878 shinies in 47,087 encounters, for a rate of 1 in 54 (with a 95% CI of 50 to 57 encounters per shiny). Aggregating should be done with caution here because the data does not definitively support the hypothesis that these Pokémon all have the same shiny rate.

The preceding table combines data from several different Research Group projects, including our regular shiny-tracking project, several Task Forces, and our ongoing project tracking the distribution of the Pokémon GO Egg pool. One striking observation is that data from the Egg hatching project consistently yields slightly more conservative shiny rates than the other projects do.

In addition, nearly all of the data on Baby Pokémon comes from the Egg pool project. There are several good reasons to suspect that data from the Egg pool project might be more accurate than data from other projects. One is that Eggs are generally easier to track than wild encounters; there are no issues with fleeing, shiny-checking, different forms, or GO+ usage, for example. Additionally, since the primary purpose of the Egg project is to track the distribution of species in the Egg pool rather than to track shiny rates, this project may be less susceptible than others to reporting bias. Alternatively, there is the potential for Researchers to forget to note that some hatches were shiny, leading to the opposite bias. Finally, data submission for the Egg pool project is restricted to more veteran researchers.

What can we make of all this? The simplest hypothesis is that all of these Pokémon do, in fact, share the same shiny rate. If the Egg pool data is indeed more accurate than other data, then this rate may even be the same as the Permaboost rate. More complicated hypotheses could involve different shiny rates for different categories, some tweaking to the shiny rate over time, slightly different shiny rates for different encounter types, or perhaps even some variability within the categories. For example, our observed shiny rate for Bonsly is much higher than that of other baby Pokémon, although with substantial data on 11 different baby Pokémon, it’s not necessarily surprising that we might see an outlier. (We have too little data on Happiny, Mime Jr., and Riolu to draw any conclusions about them.) Similarly, it remains possible that the shiny rate for Klink might be higher than that of Absol, Mawile, and Shinx.

---

Limited Research Events

Niantic has held four three-hour Limited Research events, featuring Feebas, Clamperl, Lotad, and most recently Minccino. Notoriously, the shiny rate for Feebas was changed from roughly 1 in 40 to roughly 1 in 20 after negative feedback from trainers in Europe, the Middle East, Africa, and India. In the three subsequent events, our Researchers recorded the following observations:

Given what we understand about the way that shiny rates are re-used, we think it’s likely that the shiny rates for Clamperl and Minccino were the same as the Raid Day rate, while Lotad most likely used the Legendary Raid rate.

The Lotad event also coincided with the release of shiny Normal Castform. Our Researchers encountered 12,934 Normal Castform during the event, of which 829 were shiny; this is a rate of 1 in 15.6, with a 95% CI of 14.6 to 16.7 encounters per shiny.

---

Rates With Limited Data

We close by listing shiny rates in several additional categories which either are unique (e.g. Meltan, Spinda) and/or for which we have data that is too limited to draw any significant conclusions.

The Costumed Egg/Raid category includes the Halloween 2019 costumed Kanto starters, as well as Stantler with bells and the more recent Raticate, Wobbuffet, and Wurmple with party hats. Our data for each of these species individually is too limited to draw any conclusions about whether they are all actually at the same rate.

Gyarados data is from the 2020 Lunar New Year event, and Chansey data is from Valentine’s Day 2020 (its shiny release event).

Summary

Our review of Raid, Egg, and other shiny data reveals that Niantic has used a consistent shiny rate of approximately 1 in 20 for Legendary Raids, and approximately 1 in 10 for three-hour Raid events. Three-hour Limited Research events have used both of these rates.

The picture is less clear-cut for Raid/Egg-exclusive Pokémon. Baby Pokémon appear — at least for the most part — to use a rate of approximately 1 in 58 (with a 95% confidence interval of 53 to 64 encounters per shiny). This rate may very well be the same as the Permaboost rate used by at least a dozen species (such as Sneasel) with shiny rates permanently higher than the base rate. Our observed rates for other Raid/Egg exclusive Pokémon, such as Absol, Mawile, Shinx, and various Alolans are higher; but whether this is because their shiny rates are actually higher, or due to bias and/or luck in data collection, remains a key unanswered question.

Finally, our data for certain other encounter types (such as Spinda in Field Research and Shadow Pokémon from Team Rocket leader battles) is sufficient to conclude that their shiny rates are higher than the base rate, but too preliminary to categorize their rates precisely.

We hope this information aids you in your hunt for shiny Pokémon. Keep an eye out for the third and final installment in the coming weeks!

---

Appendix 1 — Detailed Raid/Egg-exclusive data

Appendix 2 — Footnotes

¹ Absol, Alolan Meowth and Vulpix have also been available in quests, and Absol, Mawile, and Alolan Sandshrew and Vulpix had limited wild availability during an event.