*** June Competition Winners: Well done to Le Male (NAPS), Coco69 (KO Cup), Saddlesore (Royal Ascot) & Like2Fish (Poker) ***
*** New Last Man Standing Competition - Win up to £1500 Annually - See Competitions Forum ***


New Members
  • Content Count

  • Joined

  • Last visited

1 Follower

About perkon

  • Rank

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. I also observed the idea which is described in the original post. When I built up my excel stat sheets, I called this stat "streak stop", which is similar this "tiers" thing. My "solution" : In my system the teams are ranked by ELO points, so the league tables are sorted by ELO points. I divided this power ranking league tables into three parts. ("strong-middle-weak" teams). In each part I observed the WDL streaks' averages contiunally through the season in total and in H/A as well. And if one team store one result (W-D-L) up (reaching or pass the observed streak's averages of this result), this "streak stop" stat signs this warning call and we should bet against this result (maybe a "surprise" result come up from this). Of course these "streak-averages" are just random patterns and no more. And this "patterns" breaking are not a principle, rather a tendency to do so. In nutshell this should be the thing. For example this WDL streak avereages in English Premier League 2017/18 after 31 round (in april 2018): Total: 1. (1-6 ELO positions): 6-2-2 (WDL) 2. (7-14 ELO positions): 3-2-3 (WDL) 3. (15-20 ELO positions): 2-3-5 (WDL) Home averages: 1. 7-2-1 2. 3-2-2 3. 2-2-3 Away averages: 1.4-2-2 2.2-2-4 3.1-2-4
  2. Thank you liero! Slowly two years I am constantly developing the basic table, involving many new, but basic stats and datas. I think for now the basic stats and structures are ready to gather from week to week. So, sadly there are no huge historical database. But from now every weekend I create and collect the stats and the database will be bigger and bigger. Otherwise theoretically the basic table is suited to make stats from any time from the past, if the historical results and odds are available.
  3. Ok then! Here are the previous two weekend results filters: Feel free to comment your opnions or profitable filtered situations!
  4. Ok! Good luck for your experiment. I will reupload the sheets if it will be allowed to publish here. It contains a "Notes" part, a brief summary. The databse is just contains the previous weeks' results but week after week it will be bigger. My observation: I think the poisson distribution is fairly accurate with low scoring games where ELO ratings differences' is small between teams. (ok, the pattern is irrelevant small for now...)
  5. Well, I fully understand your concerns about it and in global you're totally right but if you take notice of my side, I have no other option but write a download link or site reference, cause there is no another solution to display here such a big (hundreds of datas) information. I know for the greater goods we may punish the confinding members. You alluded to my end game: Well, as I mentioned earlier I like to observe basic stats, looking for relationships to find some (maybe) profitable situations, so my final goal is to share this database to others and the possibility to everyone of searching this kind of situations with this filtering process cause there are millions of different filtering we made, and this is a very time consuming thing, mops up a lot of working hours (even though excel macro quicken the process). Of course this sheet is developing constantly.(for nearly a year). So far the undermentioned stats could be filtered: ELO Points, Forms, Form Bends, Streaks, Result Odds, Ante post odds, 1X2 and goal tips by stats,Last 5 matches sequences, Team balances, League balances, Team positons, Fair odds, Poisson tips, Value%, etc. An example: I trying to find possible draw matches with a three filtered way: 1. Set the ELO point differences down. between: -50 and 50. 2. Set the 3 most probable result by poisson at first place to 1-1. 3. Set the Home Team at Home Draw balance >30%. After this filtering process in the weekend matches (2019.01.12.-01.14.) there are 7 games and from these games there are 5 drawn games with 136% yield. (1 0-0, and 4 1-1). It's just a small example and I know it's a constant trials or errors, but more than nothing.
  6. Seriously don't understand why won't allowed to share free stats and datas in this complex ways (filter by more than 400 conditons). Personally don't just advertise any site, but don't have any chance to share my excel. (no download links, no site reference.) In this excel sheets I have a lot of working hours and want to share with the betting community for free (of course just to who interested). Anyway Trefik, Cgmbet (paid services), and rdata allowed to advertise his sites. Smells like kind of a double standard... How this direction helps the betting community? Or is there a rule here to use for exapmle ELO ratings, tips only from this site or something...? No offense, just curious.
  7. Ok! Ready for the week-end with the excel stats and tips. Unfortunately there is no option to post a link for the sheet, so I will stay with the screenshot. If anyone interested contact me!
  8. Ok then! So in the name of the principle of the presumption of innocence, in my second post I post a screenshot about a 2 filtered situation from the previous week seeking probable draw matches. 1.filter: matches where most probable result 1-1 calculated by poisson dist. 2. Total points (ELO and form): between -50 and 50. It was a profitable filtering process for betting the draw (53% yield). It's just a small example and with this free excel tool there are a lot of options (according to personal choices) to filter some maybe profitable situations. I know the filtering process is just trials and errors but with this we simply controls or double-checks our premises about betting with the help of stats and datas. Don't know why protect members from free data...
  9. Hello everyone! As a stat-freak I created my own stat based system in excel sheets, which contains multiple stat datas, streaks, balances, positions, odds, ELO points, poisson calculated tips and other tips. It's only for soccer and for national championships and for the main betting (1X2 and goals categories) categories. Making the stats for the week-end games, it's a very time consuming thing... but after the week-end matches, I created also a results analysis in excel. In this analysis there are more than 400 conditions / filter columns in which we will find some interesting relationships. In this analysis after the filtering process there are hit rate % and yield %. So the link for the sheet (previous week-end games analysis):