Spam has become much easier to avoid on inbound mail at some point. I turned off all Bayesian/heuristic based spam filtering around 2015 and now just check SPF and DKIM, and have a fake first MX record. No "Junk" folder, everything that passes the checks goes to the inbox. I get maybe one spam mail every week or so, which I just delete.

Could you please elaborate on the "..., and have a fake first MX record" ?

Legitimate SMTP servers will try your domain's MX records one by one, in order of priority, until they reach one that accepts the message. Spammers' scripts usually don't bother, they just try the first one and move on to the next address on their list.

Of course, this is not 100% reliable, as it's not too difficult for spammers to adapt and improve their scripts. Of course, vast majority of spammers are either not sophisticated enough, or do not care enough to do so, so if you don't mind your incoming mail to be slightly delayed, it's kind of a low-hanging fruit, as it cuts off a huge amount of low-effort spammers.

Not GP, but I think what they meant is having a MX record with a higher priority pointing to an unroutable IP

    blackhole  IN A 240.0.0.1

    @ MX 10 blackhole.example.com
    @ MX 20 mail.example.com

Yup, exactly, for the reasons described in the sibling comment to yours.

Doing it this way doesn't even delay mail much most of the time; many legitimate MTAs connect immediately to the priority-20 MX after failing to connect to the priority-10 one.

Yes, receiving is very easy. Just doing that has a lot of value because now can't be arbitrarily cut off from your internet identity by gmail/et.al. for no reason.

So a easy way to get started is to receive everything directly and use a commercial (often with low-volume free tier service) relay for outbound until you get comfortable enough to remove the training wheels. (Or never remove them, that's a legit choice as well.)

> filtering out the spam (and only the spam) from your inbound email is harder.

I don't find that at all. Filtering spam is the easiest part. All I do is if SPF doesn't match, goes to spam folder. Beyond that, apply a bayesian filter.

I get no false positives and the spam that gets through to my inbox can be counted on one hand per quarter. Basically none.

That's yet another benefit of self hosting, since my bayesian filter is trained on my personal email specifically, it tends to become very good. Unlike generic gmail filters for example, where there'll always be some mail that ends up in spam no matter how many hundreds of times you mark it not-spam.

It's easy to block all spam, if you don't mind also blocking legitimate mail. What's harder is being confident you're not blocking any legitimate mail. I'm glad you've managed it :).

I have Bayesian filtering behind DMARC checks and a few other DNS and RBL checks. Still not perfect -- for all folk talk of the difficulties getting mail accepted by the big providers, a surprising number of companies still don't have matching forward and reverse DNS records :P.

One principle I follow is that I don't drop anything. Every message will either be rejected or delivered, with a narrow band of SpamAssassin scores that delivers to a spam folder.

> So a easy way to get started is to receive everything directly and use a commercial (often with low-volume free tier service) relay for outbound --- I used to do that. Did it for almost a decade, in fact. Then the spam filterers decided any message with mismatched send and receive routes was spam.

Use aliases?