Well Sony have spooke that PS4P had a good mix of new customer and upgrades one, I believe it was like 40% upgrades. Don't have the source, would need to google it.

oh wow, its all about context man, october was a month switch was fully stocked every where, and had one of the most hyped and highest rated mario games of all time. mario sold over 900k in npd in 2 days, yet sales didn't sky rocket for hardware so it was very surprising, now look at september for ps4, it launched with a none exclusive big (destiny) game, it did 538k. the fact that you did a graph for first october's is cute though.

Hell, Mario Kart Wii is #96 right now in the hourly best-sellers chart on Amazon, and New SMB Wii is at #146. They just keep selling despite closing in on a decade old. No wonder they both pulled over 30 million copies.

A lot of people grossly overestimated how well the Switch would do. I've seen a lot of 400k+ predictions both here and on ResetEra, some even over 500k and 600k. I have no idea why in the hell anyone would think the Switch would sell that damn much in a 4-week month, given that at best Super Mario has a modest effect on hardware sales. Someone must have been misinterpreting some prior LTD sales estimate. NPD had the Switch at a bit over 2.32M at the end of September, and that news report from a couple of weeks ago said the Switch was at "over 2.6 million" at the end of October, meaning the Switch would have to have sold somewhere in the 275k to 375k range in October, with a reasonable assumption that the actual number was at the low to mid end of the range. As it turns out, it was towards the lower end. Anybody that estimated anything 400k or more was either unaware of that news report, or had no idea where the Switch was at the end of September.

My own prediction was one of the most conservative out of all the estimates I've seen, overestimating the actually tally by only about 11-13% (assuming sales of 295k ±3k), not too far outside any reasonable margin of error (and when we're just guessing, a MoE of 10% isn't unreasonable). I was assuming the first three weeks of October would retain the 63k/week average of September, with SMO driving the fourth week to about 2 to 2.5 times the aforementioned 63k/week average, putting it in the 315 to 346.5k range. I split the difference down the middle and guessed 330k. Either SMO didn't give as big of a boost as I thought, or weekly sales for the first three weeks declined a bit from the September average.

So, while a lot of people may have been far, far off the mark, the sales themselves when viewed purely on their own relative to other systems and the Switch's own prior sales are not unimpressive. They are actually quite good, even if SMO wasn't a massive system-seller.

It's quite a big percentage of people double dipping (I'm not saying this is bad, it is actually great for Sony). But it's interesting to see the late adopters also getting it. I guess a lot of people don't mind paying 350-400 on a console. Increased 4K adoption is probably the reason, even late costumers seem inclined to jump ship to the most powerful unit.

MHW keeping it lead over DQ XI. I have little doubt now that it will open lower than DQ XI

Kiwani 2 also accelerating. Should do around 120 points and 150k opening. Just my opinion

Wrong thread?

And "unimpressive" gives a negative connotation, as if the sales were lackluster or average at best or otherwise worthy of little more than a "meh" reaction, and "meh" is not the reaction one should have for these level of sales. By any objective standard, 290-300k for a 4-week non-holiday month is very good, exceptional even. Excepting Februaries and Marches (because tax season boost), the PS4 has so far only exceeded a 70k/week average for a non-holiday month four times: Sept. & Oct. 2014 (Destiny effect), June 2015 (Arkham Knight bundle), and June 2017 ($249 gold PS4).

It's not my fault people were grossly overestimating the Switch's sales potential for October. Yes, I did fail to realize that most of the predictions were made before the news of 2.6M LTD broke (I didn't bother checking dates in my prior post, so total mea culpa there), but a lot were made after that fact, so not everyone has that excuse. In any case, let's see what a 400+k prediction implies. Given what it sold in September (about 63k/week avg.), we can assume that at best the Switch sold around 190k for the first three weeks of October. That means to have sold over 400k, it needed to sell at least 210k in SMO week. That's a lot for a single week. Hell, that'd be an okay month by itself.  500+k would require over 310k. If we assume VGC isn't too far off the mark in its weekly hardware sales (±10% MoE at most), that would make SMO the single-biggest system-seller of the past decade, with the possible exception of Mario Kart Wii. In Japan the Switch only managed 126.7k that week, and the Switch's sales haven't been massively different than what they are in the U.S. It ought to have sold a good bit over 100k SMO week in the U.S., which is a not-bad-but-not-great week for a week with a major game, but we shouldn't expect a massively huge boost from a Super Mario game.

Given what we know of Super Mario's system-selling capabilities, or rather lack thereof, the idea that it could have generated one of the best non-holiday weeks ever, if not potentially the best in years, is patently ridiculous. If people have unrealistic expectations for how well something will sell, we shouldn't consider that if sales fail to meet those expectations that they are "unimpressive" or otherwise imply that they're not good. It's a strike against the predictors, not the Switch. I honestly considered 330k to be a very optimistic prediction. 400k? Unrealistic. 500k? Ridiculous. 600k? Had to be a joke.

And "unimpressive" gives a negative connotation, as if the sales were lackluster or average at best or otherwise worthy of little more than a "meh" reaction, and "meh" is not the reaction one should have for these level of sales. By any objective standard, 290-300k for a 4-week non-holiday month is very good, exceptional even. Excepting Februaries and Marches (because tax season boost), the PS4 has so far only exceeded a 70k/week average for a non-holiday month four times: Sept. & Oct. 2014 (Destiny effect), June 2015 (Arkham Knight bundle), and June 2017 ($249 gold PS4).

It's not my fault people were grossly overestimating the Switch's sales potential for October. Yes, I did fail to realize that most of the predictions were made before the news of 2.6M LTD broke (I didn't bother checking dates in my prior post, so total mea culpa there), but a lot were made after that fact, so not everyone has that excuse. In any case, let's see what a 400+k prediction implies. Given what it sold in September (about 63k/week avg.), we can assume that at best the Switch sold around 190k for the first three weeks of October. That means to have sold over 400k, it needed to sell at least 210k in SMO week. That's a lot for a single week. Hell, that'd be an okay month by itself.  500+k would require over 310k. If we assume VGC isn't too far off the mark in its weekly hardware sales (±10% MoE at most), that would make SMO the single-biggest system-seller of the past decade, with the possible exception of Mario Kart Wii. In Japan the Switch only managed 126.7k that week, and the Switch's sales haven't been massively different than what they are in the U.S. It ought to have sold a good bit over 100k SMO week in the U.S., which is a not-bad-but-not-great week for a week with a major game, but we shouldn't expect a massively huge boost from a Super Mario game.

Given what we know of Super Mario's system-selling capabilities, or rather lack thereof, the idea that it could have generated one of the best non-holiday weeks ever, if not potentially the best in years, is patently ridiculous. If people have unrealistic expectations for how well something will sell, we shouldn't consider that if sales fail to meet those expectations that they are "unimpressive" or otherwise imply that they're not good. It's a strike against the predictors, not the Switch. I honestly considered 330k to be a very optimistic prediction. 400k? Unrealistic. 500k? Ridiculous. 600k? Had to be a joke.

Well, eight technically, possibly nine, assuming a minimum of 290k (72.5k/week) for the Switch's October. But five of those were during tax season, and February and March are usually the two best non-holiday months because people get that big check and buy big-ticket items like game consoles. The Switch has yet to have a tax season, which is why I excluded Feb. & March from the list in the previous post. Assuming Nintendo keeps the thing in stock after the holidays, the Switch ought to handily clear 800k for Q1, maybe even 900k (I'd say 200-250k in Jan., 350-400k in February, and 250-300k in March). Even a million isn't out of the question, though that's an extremely optimistic scenario.

The ninth potential month for the PS4 to surpass the Switch's per-week average for October was June 2015, Arkham Knight month, which was a statistical tie. We're not sure what the exact number was for the Switch, but if it was over 292k it beats the PS4's per-week average for June 2015, if only slightly.