Last month we published a full month of drop-timing data and ended it with a hedge: the pattern moves, so treat the percentages as directional. July gave us a second full month to check that against — and the honest answer turned out to be more interesting than "it moved."
The aggregate chart changed a lot. The individual retailers barely changed at all.
We bucketed the same two kinds of events by day of week and hour of day (US Eastern), using the same definitions as June so the two months are directly comparable:
- Confirmed retail restocks — the moments we detected buyable inventory and fired an alert (107 events).
- Community drop signals — drop-relevant posts surfaced from Reddit and Twitter/X (254 events).
That's 361 timestamped drop events across July 1–30, 2026, up from 243 in June.
TL;DR:
- Walmart is a Wednesday-night retailer. Not mostly — entirely. All 30 of 30 Walmart restocks we caught in July landed on a Wednesday, and all 30 fell between 9pm and midnight ET. Twenty-four of them were in the 9pm hour alone.
- Target is a Tuesday/Friday overnight retailer. 54 of 55 Target restocks fired between 3am and 5am ET, and they only happened on two weekdays: Friday (37) and Tuesday (17).
- Amazon has no clock. Its 21 restocks were scattered across all seven days and nearly every hour. It's the one retailer where "when should I watch?" has no useful answer.
- The weekend is dead, and getting deader. Saturday and Sunday together held under 5% of all activity, and just 4 of 107 confirmed restocks. June's weekend share was 12%.
- Nothing buyable appeared between 3pm and 8pm ET all month. Zero of 107. That's the safest five-hour stretch of the day to put your phone down.
Three retailers, three clocks
This is the finding worth the whole post. Instead of one blended "drop time," July shows three retailers each running a schedule of their own — and two of them are close to mechanical.
| Retailer | Restocks | The clock | Days |
|---|---|---|---|
| Target | 55 | 3am–5am ET (54 of 55) | Friday (37), Tuesday (17) |
| Walmart | 30 | 9pm–midnight ET (30 of 30) | Wednesday only |
| Amazon | 21 | none — spread across 11 different hours | all seven days |
| Macy's | 1 | — | — |
Now the part that makes this credible rather than a fluke: the same three clocks are in June's data too, at a quarter of the volume.
| Retailer | June | July |
|---|---|---|
| Target | 12 of 13 at 3am–5am ET; Friday 9, Tuesday 4 | 54 of 55 at 3am–5am ET; Friday 37, Tuesday 17 |
| Walmart | 5 of 6 on a Wednesday night | 30 of 30 on Wednesday, 9pm–midnight |
| Amazon | every day, every hour | every day, every hour |
Two independent months, same shape, four times the sample. In June we called Wednesday evening "the Walmart window" off six data points and hedged it. Thirty for thirty is not a hedge anymore.
Drops by day of week
All 361 events:
| Day | Events | Share |
|---|---|---|
| Monday | 30 | 8% |
| Tuesday | 55 | 15% |
| Wednesday | 132 | 37% |
| Thursday | 38 | 10% |
| Friday | 89 | 25% |
| Saturday | 8 | 2% |
| Sunday | 9 | 3% |
Wednesday and Friday together carry 221 of the month's 361 events — 61%. Add Tuesday and three days hold more than three-quarters of everything that happened in July.
The weekend barely registers. Eight Saturday events and nine Sunday events out of 361 — and on the confirmed-restock side, where it actually matters, the entire weekend produced four buyable restocks in a month. If you have been giving up Saturday mornings to refreshing product pages, July says stop.
Drops by hour of day (US Eastern)
| Window (ET) | Events | Share | June |
|---|---|---|---|
| Overnight (12am–5am) | 124 | 34% | 20% |
| Late morning (6am–11am) | 45 | 12% | 24% |
| Midday (12pm–2pm) | 59 | 16% | 24% |
| Afternoon (3pm–5pm) | 14 | 4% | 11% |
| Evening (6pm–11pm) | 119 | 33% | 21% |
The day collapsed into two windows. Overnight and evening went from 41% of activity in June to 67% in July, and every other part of the day shrank.
Hour by hour, one bar dominates:
ET hour events (one # ≈ 2 events)
03:00 ###################### (44)
04:00 ######################### (49)
05:00 ########### (21)
11:00 ############## (28)
12:00 ######## (16)
13:00 ############# (25)
14:00 ######### (18)
21:00 #################################### (72) ← busiest hour of the month
22:00 ############# (25)
23:00 ###### (11)
9pm ET was the single busiest hour of July, with 72 events — more than twice the size of June's busiest hour. That is Walmart's Wednesday window plus the community reacting to it in real time.
Narrow it to confirmed restocks only — the moments inventory was actually buyable — and the concentration is starker:
- 3am–5am ET: 55 of 107 restocks (51%)
- 9pm–11pm ET: 39 of 107 restocks (36%)
- Everything else, all 18 remaining hours combined: 13 restocks
- 3pm–8pm ET: zero
Nearly nine out of ten buyable moments in July landed inside two four-hour windows.
Why the chart moved when the retailers didn't
June's aggregate said Wednesday, midday-heavy, with restocks scattered around the clock. July's says Wednesday and Friday, overnight-and-evening, with a dead middle. Same question, two very different pictures, one month apart.
The retailers didn't change. The mix did.
In June, Amazon was 64% of all confirmed restocks (37 of 58) — and Amazon is the one retailer with no schedule, so a month it dominates looks smeared across the clock. In July, Amazon fell to 20% (21 of 107) while Target and Walmart went from a third of the month's restocks to 79% of them. Two retailers with hard schedules replaced one retailer with none, and the aggregate chart snapped into their shape.
That reframes the advice we gave in June. It isn't that drop timing is unstable — it's that an aggregate across retailers is unstable, because it's a weighted average of clocks that don't move. Which retailer is doing the restocking in a given month decides what the blended chart looks like.
There was plenty for them to restock. July ran 48 distinct products across the Pitch Black launch, Chaos Rising, Ascended Heroes and Perfect Order — a heavy month at the two retailers that keep the tightest schedules.
What to do with this
- Wednesday, 9pm ET: that's Walmart. Thirty for thirty in July, five of six in June. If you only defend one hour a week, it's this one.
- Tuesday and Friday, 3am–5am ET: that's Target. Inconvenient, which is exactly why it's worth an alert instead of an alarm clock.
- For Amazon, don't bother with a clock. It restocked on every day of the week at eleven different hours. Timing is not the lever there; watching the price is.
- Write off 3pm–8pm ET. Not one buyable restock landed in that window in the entire month.
- Write off the weekend too. Four restocks across nine weekend days.
- Judge a month by which retailers are active, not by last month's chart. A month where Amazon does most of the restocking will look nothing like a month where Target and Walmart do.
- Or skip all of it and let alerts do the watching. The reason we can publish a 4am Friday number is that nobody here is awake for it either.
We'll keep logging and publish this again next month. Two months in, the useful conclusion isn't a single drop time — it's that the big retailers each keep one, and they keep it consistently enough to plan around.
Get Pokemon TCG restock alerts at alerts.bujusjujus.com →
Methodology note: based on 361 timestamped drop events from our restock-alert telemetry over July 1–30, 2026 — 107 confirmed retail restock alerts and 254 community drop signals — bucketed by day of week and hour, converted to US Eastern (EDT). A small number of evergreen search-feed refreshes are set aside rather than counted as distinct restocks, and the month's final day is not in the sample. June's figures are recomputed here by the same method for comparison. 107 restocks over 30 days is a real but small sample: the per-retailer day-and-hour patterns are strong enough to state plainly, but thinner slices of it are not. Previous timing piece: When Do Pokemon TCG Drops Actually Happen?. Companion piece: The State of Buying Pokemon TCG at MSRP — July 2026.