WAIT.

You think Switch only outsold PS4 by 20k this month? How? Last month the gap was 10k, you think that Mario Odyssey only increased the gap by 10k units?

PS4 will of course see a drop, now, October is a 4 weeks month compared to September 5 weeks. Even if PS4 had the same avg in weekly sales, which seem kinda unlucky since last month was Destiny and NBA which were bigger than AC Origins and Shadow of War, PS4 would be at 240k this month.

At best, PS4 got outsold by 50k.

"Cute"? Don't get smart with me, son. I know you've been here for only a week, but I've been doing graphs like these for a long time now. That's kinda what I'm known for here. I'm the "graphs guy." If you're going to get snide with me, I'm just going to ignore you. If you want me to continue to debate you, then watch your tone.

"Cute"? Don't get smart with me, son. I know you've been here for only a week, but I've been doing graphs like these for a long time now. That's kinda what I'm known for here. I'm the "graphs guy." If you're going to get snide with me, I'm just going to ignore you. If you want me to continue to debate you, then watch your tone.

I mentioned the effect of software on hardware sales before. The effects are less obvious with monthly tracking (as NPD does) than with weekly tracking (as the Japanese trackers do), especially since Japan already appears to be more responsive to software than the U.S. A game can boost weekly sales by double, but the net effect for the whole month is going to be a lot smaller. Say a system's been averaging 50k/week, and a notable new game boosts sales to 100k for the week it's released, while nothing big released the previous month. Assuming both the month the game was released and the prior month were both 4-week months, that's 250k for the month the game was released and 200k for the prior month. While the game yielded a 100% boost to sales for the week it was released, it yielded only a 25% month-over-month increase. If the prior month was a 5-week month instead, then the system would have sold 250k in both months despite the game in the second month boosting hardware sales to double baseline the week it was released.

The Switch's per-week average for October in the U.S. was 16% higher than in September. Considering the effects SMO had on Japanese sales, we can surmise that the game gave a pretty good boost to sales the week the game was released in the U.S. as well. How big, we don't know, and the effects on the month of October as a whole will be smaller than the effect on sales in the week the game came out. Incidentally, looking at Japanese sales, SMO gave a good boost to sales, but the net effect was to make October (Media Create Weeks 40-43) have a weekly average only 15.5% larger than in September (MC Weeks 35-39).

Also, no two system-sellers were born the same. A rare few have massive effects on sales, sometimes lasting several weeks or more (Japanese data suggests that the vast majority of system-sellers help for a week, maybe two at most, and we can probably assume that's the norm elsewhere), while most have a more modest effect on sales. The PS4 got an absolutely massive spike in the U.S. in Sept. 2014 from Destiny, one of the biggest system-sellers ever. That month, the PS4 averaged a staggering 126.5% increase in weekly average sales from August, far beyond anything else we've seen on a home console in the past two generations. The following month, the PS4 still averaged well above the April-Aug. baseline, suggesting Destiny was still having a residual effect on sales. Nothing new was released in October 2014 that could have sustained sales, and no price cuts or notable bundles/promotions were issued that month, leaving Destiny as the only probable explanation. Other extremely big system-sellers in the past decade include Metal Gear Solid 4 for the PS3, Pokemon X/Y for the 3DS, and Halo 3 for the 360 (except in Japan). Those games moved a ton of hardware.

But most system sellers do a lot less, and, as I talked about earlier, with NPD tracking sales on a monthly basis, the vast majority of them don't have an effect on sales that can be determined from the data. Mario Odyssey was a system-seller, but not a massive one. It provided similar month-over-month boosts in both the U.S. and Japan, with the latter showing a solid boost for the final week of October, but not one so large as to cause massive month-over-month growth, and it's a safe assumption that about the same thing happened in the U.S. SMO not being a Pokemon-caliber system-seller for Nintendo is not a strike against the system.

In fact, 3D Super Mario games are, surprisingly, typically not known for moving a ton of hardware by themselves. Sunshine did squat to help the GameCube in the U.S., and helped only a little bit in Japan. Neither Galaxy game had much of an effect as far as I can tell. 3D Land gave the 3DS a decent boost in Japan, but nothing massive, and 3D World didn't do much of anything to help. SMO being a relatively modest system-seller is not surprising in the least.

Wii was on Fire! Great sales for Switch! 

"Cute"? Don't get smart with me, son. I know you've been here for only a week, but I've been doing graphs like these for a long time now. That's kinda what I'm known for here. I'm the "graphs guy." If you're going to get snide with me, I'm just going to ignore you. If you want me to continue to debate you, then watch your tone.

I mentioned the effect of software on hardware sales before. The effects are less obvious with monthly tracking (as NPD does) than with weekly tracking (as the Japanese trackers do), especially since Japan already appears to be more responsive to software than the U.S. A game can boost weekly sales by double, but the net effect for the whole month is going to be a lot smaller. Say a system's been averaging 50k/week, and a notable new game boosts sales to 100k for the week it's released, while nothing big released the previous month. Assuming both the month the game was released and the prior month were both 4-week months, that's 250k for the month the game was released and 200k for the prior month. While the game yielded a 100% boost to sales for the week it was released, it yielded only a 25% month-over-month increase. If the prior month was a 5-week month instead, then the system would have sold 250k in both months despite the game in the second month boosting hardware sales to double baseline the week it was released.

The Switch's per-week average for October in the U.S. was 16% higher than in September. Considering the effects SMO had on Japanese sales, we can surmise that the game gave a pretty good boost to sales the week the game was released in the U.S. as well. How big, we don't know, and the effects on the month of October as a whole will be smaller than the effect on sales in the week the game came out. Incidentally, looking at Japanese sales, SMO gave a good boost to sales, but the net effect was to make October (Media Create Weeks 40-43) have a weekly average only 15.5% larger than in September (MC Weeks 35-39).

Also, no two system-sellers were born the same. A rare few have massive effects on sales, sometimes lasting several weeks or more (Japanese data suggests that the vast majority of system-sellers help for a week, maybe two at most, and we can probably assume that's the norm elsewhere), while most have a more modest effect on sales. The PS4 got an absolutely massive spike in the U.S. in Sept. 2014 from Destiny, one of the biggest system-sellers ever. That month, the PS4 averaged a staggering 126.5% increase in weekly average sales from August, far beyond anything else we've seen on a home console in the past two generations. The following month, the PS4 still averaged well above the April-Aug. baseline, suggesting Destiny was still having a residual effect on sales. Nothing new was released in October 2014 that could have sustained sales, and no price cuts or notable bundles/promotions were issued that month, leaving Destiny as the only probable explanation. Other extremely big system-sellers in the past decade include Metal Gear Solid 4 for the PS3, Pokemon X/Y for the 3DS, and Halo 3 for the 360 (except in Japan). Those games moved a ton of hardware.

But most system sellers do a lot less, and, as I talked about earlier, with NPD tracking sales on a monthly basis, the vast majority of them don't have an effect on sales that can be determined from the data. Mario Odyssey was a system-seller, but not a massive one. It provided similar month-over-month boosts in both the U.S. and Japan, with the latter showing a solid boost for the final week of October, but not one so large as to cause massive month-over-month growth, and it's a safe assumption that about the same thing happened in the U.S. SMO not being a Pokemon-caliber system-seller for Nintendo is not a strike against the system.

In fact, 3D Super Mario games are, surprisingly, typically not known for moving a ton of hardware by themselves. Sunshine did squat to help the GameCube in the U.S., and helped only a little bit in Japan. Neither Galaxy game had much of an effect as far as I can tell. 3D Land gave the 3DS a decent boost in Japan, but nothing massive, and 3D World didn't do much of anything to help. SMO being a relatively modest system-seller is not surprising in the least.

Please accept that anything Madword says from now on also represents what I think. 

First off, who cares what everyone else was predicting? It's absolutely irrelevant to whether or not the actual numbers are strong. Same for SMO's Metascore. A high review average does not a system-seller make. Most of the best-reviewed games ever didn't shift an appreciable amount of units at all. In fact, you've made more irrelevant points than I care to discuss, like that whole "global shipments are the only thing that matters" comment you made the other day. It's a complete non-sequitur, having nothing to do with the subject of whether it's appropriate to align two systems by launch month if one launches in November and the other in March.

Second, if you're going to give me a hard time and impugn my character by accusing me of pro-Nintendo bias and insinuating that I'm deliberately cherry-picking data to "make Nintendo look good," then we're done. I'm under no obligation to put up with your crap. If you must know, I've been doing charts like that one since at least February 2016, and I started doing that particular style because the all-months chart I had been using got to look like a tangled mess after so many data points were added (a line graph covering 26 months and nine systems isn't going to be easy to make presentable and readable). It had nothing to do with making any system look good or otherwise trying to manipulate the data to provide a favorable image of any system. All of my charts (and I have a lot of them) have always been about presenting the data in an easily-digestible manner to illustrate trends, relative performances, the effects of software and price cuts, and other points of interest in the sales data, from the basic to in-depth.

So, you have two options: Either A) apologize for your insinuations about my intentions, or B) don't bother replying to me again, because you'll be wasting your time.

Seeing as the only/closest numbers we have for Odyssey's first 3 days is over 1m, I'll give you the win. Gg.