![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Words from the Major-General
Mar. 30th, 2008 12:17 pmMy Office Junk-Mail filter does some odd things (as indeed does my BT filter -- I really should check that more often). THis morning I had two e-mails. The "Bet-At-Home Get-5-Euros-FREE!!!" email got through without a murmur. Meanwhile, Pete Doubleday's response to the previous post (all responses also come through to me as e-mails) was consigned to the Junk-Mail section.
This had never happened before with an LJ response, not even the Russia-sourced pron ography links. Clearly a surfeit of the word "Ubuntu" and some other factors were enough to convince Office Spam filter that this communication was a pile of junk.
I get this problem occasionally with my newsletter. Given the nature of the publication, I often mention large sums of money. I also often refer to various US attorneys-general. Now, people who produce spam-detecting algorithms don't know much about insurance, or the legal system, or, indeed, much at all, as far as I can see. But they do get to see a lot of spam communications. They create their algorithms by analyzing these communications. The Nigerian 419 scams usually mention large sums of US dollars and the word "General" also appears quite frequently. What the people who produce spam-detecting algorithms do not do is check other niche areas of e-mailing that might produce false positives -- such as mine.
So, what happens is this. I send out the e-mail. A few hours later, Edna Thingystob at Mega-Re Bermuda e-mails me and asks where today's newsletter is. I reply that I have sent it, and that the company's spam filter has probably quarantined it. I then tell Edna Thingystob at Mega-Re bermuda that she needs to contact her IT department to see if their spam-filtering software has made a mistake. Usually, it has.
However, this will normally be the precursor of several of these e-mails, because most of the algorithms are produced by just a few companies. Messagelabs are my own bete-noire.
And I can't ask them what's wrong with a particular issue because:
(a) no-one in these companies actually understands the algorithms that have now been created. It's like a computer coming up with the answer "42", but no-one knowing how it got there.
(b) even if they did know, they wouldn't tell me, because, if I was a real spammmer, it would be very useful information to have.
This is a standard problem of "It isn't my fault" that permeates the ever-more complex society in which we live. The COO at Mega-Re of Bermuda has to do something to stop the masses of junk e-mail. The boffins at Messagelabs then take spam communications and find common threads. I am merely writing about what I write about.
The only thing which would be nice would be an automated response if an e-mail was diverted to the spam section. Then I could be a bit pro-active about it.
+++++++++++
I could do a quarterly update on the poker. This month has been hard work. I've put in about 22,000 hands because I've never been able to get going. I'm clocking up 250+ hands an hour these days, so the actual time spent playing hasn't gone up enormously. I've crawled to $850 profit for the month, the majority of it bonnuses and rakeback. Playing 22,000 hands at the tables and only breaking even is fairly dispiriting. However, I have definitely been running bad, so it's merely a matter of grinding through it.
For the quarter as a whole it's been satisfactory. January was good and February was good+. There's probably one spectacular month waiting for me this year and one horrific one.
At the beginning of the year I hoped for $15 an hour average, and I'm two bucks an hour ahead on that. My win-rate per 100 is only a fraction ahead of expectation and remains below target. I'd expected about $7 per hundred and I have a target of $10 per hundred. I'm clocking up $7.50 per hundred. The improved earn, therefore, is basically down to me getting more hands in per hour. 4-tabling (and some 5-tabling) on IP, where the games have speeded up considerably in recent months.
I've played 50,000 hands in the first quarter in 215 hours at the tables. That puts me on track for 30,000 more hands this year than I played in 2007, in roughly the same number of hours.
I'd targeted $12K and had vaguely aspired to $15K for the whole year. I'm well ahead of the first target and fractionally ahead of the second. I've had 10 winning weeks and two losing ones (where a "week" is seven consecutive days of play, rather than a calendar week). I"m averaging just over $300 a week with a standard deviation of $351 a week. Worst week was minus $500 and best week was plus $800. Worst day was minus $450 and best day was plus $460. Standard deviation per hour is clocking in at $54 -- about $35 at $100 buy in and $120 at $200 buy-in. I'm clicking in at about 4PTBB per 100 hands at $100 buy-in. $200 is 1.35PTBB per 100 hands (data consists of 18,000 hands).
I've upgraded Platinum next month on NoIQ. Betfred has been my better site all round ($13.65 per hundred over 20,000 hands), but I had a hangover of a first-deposit bonus there for the first few weeks of the year.
The regular deposit bonus on Betfred and the monthly bonus based on hands played the previous month (although you have to carry on playing the following month to earn it, so it isn't like real rakeback) currently work to give me as good a deal as I get with regular rakeback on NoIQ. However, at higher stakes and if I played more hands, the straight rakeback deal would be better.
Over the three months, I've won about 53% of my profit from rakeback and bonuses, and 47% from play at the tables. Prior to March, it was closer to 30% from the former and 70% from the latter. I'd like to average at least 60% of total profit from table play over the year.
Finally, there's about $150 in "hidden" rakeback that I haven't yet claimed. These being the small bonuses you get in Barty/Betfred/NoIQ points. That should be an end-of-year bonus of about $600.
_______________
This had never happened before with an LJ response, not even the Russia-sourced pron ography links. Clearly a surfeit of the word "Ubuntu" and some other factors were enough to convince Office Spam filter that this communication was a pile of junk.
I get this problem occasionally with my newsletter. Given the nature of the publication, I often mention large sums of money. I also often refer to various US attorneys-general. Now, people who produce spam-detecting algorithms don't know much about insurance, or the legal system, or, indeed, much at all, as far as I can see. But they do get to see a lot of spam communications. They create their algorithms by analyzing these communications. The Nigerian 419 scams usually mention large sums of US dollars and the word "General" also appears quite frequently. What the people who produce spam-detecting algorithms do not do is check other niche areas of e-mailing that might produce false positives -- such as mine.
So, what happens is this. I send out the e-mail. A few hours later, Edna Thingystob at Mega-Re Bermuda e-mails me and asks where today's newsletter is. I reply that I have sent it, and that the company's spam filter has probably quarantined it. I then tell Edna Thingystob at Mega-Re bermuda that she needs to contact her IT department to see if their spam-filtering software has made a mistake. Usually, it has.
However, this will normally be the precursor of several of these e-mails, because most of the algorithms are produced by just a few companies. Messagelabs are my own bete-noire.
And I can't ask them what's wrong with a particular issue because:
(a) no-one in these companies actually understands the algorithms that have now been created. It's like a computer coming up with the answer "42", but no-one knowing how it got there.
(b) even if they did know, they wouldn't tell me, because, if I was a real spammmer, it would be very useful information to have.
This is a standard problem of "It isn't my fault" that permeates the ever-more complex society in which we live. The COO at Mega-Re of Bermuda has to do something to stop the masses of junk e-mail. The boffins at Messagelabs then take spam communications and find common threads. I am merely writing about what I write about.
The only thing which would be nice would be an automated response if an e-mail was diverted to the spam section. Then I could be a bit pro-active about it.
+++++++++++
I could do a quarterly update on the poker. This month has been hard work. I've put in about 22,000 hands because I've never been able to get going. I'm clocking up 250+ hands an hour these days, so the actual time spent playing hasn't gone up enormously. I've crawled to $850 profit for the month, the majority of it bonnuses and rakeback. Playing 22,000 hands at the tables and only breaking even is fairly dispiriting. However, I have definitely been running bad, so it's merely a matter of grinding through it.
For the quarter as a whole it's been satisfactory. January was good and February was good+. There's probably one spectacular month waiting for me this year and one horrific one.
At the beginning of the year I hoped for $15 an hour average, and I'm two bucks an hour ahead on that. My win-rate per 100 is only a fraction ahead of expectation and remains below target. I'd expected about $7 per hundred and I have a target of $10 per hundred. I'm clocking up $7.50 per hundred. The improved earn, therefore, is basically down to me getting more hands in per hour. 4-tabling (and some 5-tabling) on IP, where the games have speeded up considerably in recent months.
I've played 50,000 hands in the first quarter in 215 hours at the tables. That puts me on track for 30,000 more hands this year than I played in 2007, in roughly the same number of hours.
I'd targeted $12K and had vaguely aspired to $15K for the whole year. I'm well ahead of the first target and fractionally ahead of the second. I've had 10 winning weeks and two losing ones (where a "week" is seven consecutive days of play, rather than a calendar week). I"m averaging just over $300 a week with a standard deviation of $351 a week. Worst week was minus $500 and best week was plus $800. Worst day was minus $450 and best day was plus $460. Standard deviation per hour is clocking in at $54 -- about $35 at $100 buy in and $120 at $200 buy-in. I'm clicking in at about 4PTBB per 100 hands at $100 buy-in. $200 is 1.35PTBB per 100 hands (data consists of 18,000 hands).
I've upgraded Platinum next month on NoIQ. Betfred has been my better site all round ($13.65 per hundred over 20,000 hands), but I had a hangover of a first-deposit bonus there for the first few weeks of the year.
The regular deposit bonus on Betfred and the monthly bonus based on hands played the previous month (although you have to carry on playing the following month to earn it, so it isn't like real rakeback) currently work to give me as good a deal as I get with regular rakeback on NoIQ. However, at higher stakes and if I played more hands, the straight rakeback deal would be better.
Over the three months, I've won about 53% of my profit from rakeback and bonuses, and 47% from play at the tables. Prior to March, it was closer to 30% from the former and 70% from the latter. I'd like to average at least 60% of total profit from table play over the year.
Finally, there's about $150 in "hidden" rakeback that I haven't yet claimed. These being the small bonuses you get in Barty/Betfred/NoIQ points. That should be an end-of-year bonus of about $600.
_______________
