Arbusers

arb12 wrote: ↑

Sun Jan 28, 2024 12:31 pm

...
The Project codenamed "Triple Play Revisited v.2.0"
...

Bayer'04 and Bayern have made significant changes to their Deutsche Bundesliga Winner odds after today's clash in some agencies. That's the visible move, but I'm waiting for more precise future reassessments for additional markets very soon, not only in that competition, so be quicker than the agencies in the next hours.

The Football Olympus, the Super Bowl LVIII encounter: Chiefs vs 49ers, is starting within a few hours.

Although I don't agree with one of the posted by the agencies' Spread lines (+1.5), in the end, some market weapons were loaded on specific positions on lines (+1.5) and (+2.5). My rationale behind that:
- I don't agree with the agencies that publish these spread lines priced at exactly those odds as they are at the moment, but that could be helpful on some occasions;
- The actual probability as of today of touching or not the quota of four touchdowns for Brock Purdy has an extremely big influence on today's Model tune-ups, let's say XXY1;
- The relative long-term stability in terms of game achievements of the duo Travis Kelce & Harrison Butker, pre-estimated as XXY2 influence on the Model.
- I have some trouble with exact pre-estimating Rashee Rice, but as a workaround believe some EMA graphics for the last periods;
- In my opinion, the last best game for the 49ers in terms of overall stability on the pitch was in December vs Washington Redskins. The problem here is that the 49ers were an extremely big favorite, so the Stats inputs aren't so reliable...
- I can't even explain to myself why the 49ers with Purdy sank vs Baltimore Ravens 19:33 but some warnings were generated;
- Statistically raised probabilities for some pre-selected market attacks in the first quarter, provided that San Francisco behaves the usual way;
- Kansas City due to its pre-estimated conditional stability (let's say YY1) and if Kelce - Butker - Rice all of them they cover at least their basic pre-estimation at some point, the Point Spread and Money Line markets would be more impressive;
- And so on.

I'd love to read the opinions of Football pundits here before Super Bowl LVIII kicks off.

Deutscher Basketball BBL,

Historical comeback for Bayern München under Pablo Laso vs Chemnitz minutes ago. What a game, OMG!!!
Absolutely the same Model tune-ups as Partizan vs Maccabi Tel Aviv in the EuroLeague Basketball held on the 12th of January, 2024, and almost absolutely the same turnaround, in both games the Favorite won by 9 points.
The only difference was that pre-match Favorite was behind by 24 points at some time (Partizan Belgrade) and 18 points (Bayern).
I am seriously thinking of specializing mainly in that beautiful kind of sport one day in the future, but there is no guarantee at all that the current Model tune-ups will continue so happily into the foreseeable future…

By the way, my Model gave me a very big unexpected error days ago - PAO Athens vs Aris Thessaloniki (Points spread market), so I'm wondering where exactly I miscalculated here...

The Hellenic Cup Final, the Basketball Superclásico Panathinaikos vs Olympiakos is going to start very soon.

The last time a very clear unconditional value in terms of published pre-match odds on some Money Line and Points Spread markets was detected by my custom Model tune-ups, was somewhere in October'23: GBL's game OLY vs PAO 54:66 and also EuroLeague's encounter PAO vs OLY 78:88 AOT.

However, the agencies have recalculated their lines to kill the hypothetical value in January'24 (PAO vs. OLY 85:80) and also before today's match.
In order to build my best trading plan for today's clash, some things were highlighted in the tune-ups:
1. Although Olympiakos hasn't been the same fortress in recent months without Vezenkov and Sloukas, it seems to me that Bartzokas struck a solid balance between the team's departments in January and February, even at the WiZink Center vs. Real. All my graphics and data cells assume certain things;
2. Sloukas' decisions today in green.
3. I wonder if Papanikolaou would be ordered more tasks and their probable implementation vs the green defenders.
4. Despite Ataman's overall excellent in-play game management, I get the impression that lately some of his canonical changes from offense to defense were awaited by the opposing coach. Very good defensive department play vs Fenerbahçe and bad vs Aris days ago, excellent in-play decisions vs Valencia and bad defensive implementations vs Saski Baskonia in January. I have no explanation for the last quarter meltdown vs Maccabi Tel Aviv in January. Therefore, I will wait not only for the Panathinaikos starting roster and the ideas for the rotation but also to see Ataman's tactical decisions today before going into many more positions. Dimitris Diamantidis is no longer playing to score some magic 3-pointers in the last minute…
5. Other observations and guesswork, of course.

Based mainly on items 4 and 1, some ideas for low-risking "middles" are arising. Additionally, probabilities for O/U could be calculated, but that's risky, due to the Overtime played between those teams in October (EuroLeague). In general, the Money Line will be so liquidity today and also the Point Spread would be low-risk manageable.
And so forth.

I would love to read some modelers' opinions for that Superclásico, if any.

EuroBasket Qualifications, Bulgaria vs Germany is going to start within a few hours.

Given that Germany are the current World Champion and the current form of both teams, the published MoneyLine odds are far from the value here (according to Kelly's Criterion).

Some of my ideas for today's trading plan are based on:
- Created some different appropriate limits on different Point Spread lines, aiming risk lowering of the "middles" attack;
- Total Points market - an equation is almost created on the basis of lower risk than usual for some pre-selected lines; the only problem here is I have some uncertainties about the rotation on the pitch today;
- Germany's dominance varies under Herbert's governance, see the full video not only of the match vs Montenegro days ago but also of the World Cup;
- The Bulgarian team under Barchovski historically has statistically clear time-limited moments of hard resistance and also time-limited heavy attacks (varying efficiency of points percentage, depending on the opponents' defense, though);
- Sasha Vezenkov (Olympiakos and Sacramento Kings) is absent again;
That's a huge problem for Bulgaria. I saw the special sparing attitude from Barchovski to Vezenkov, aiming to preserve his gaming freshness for the late attacks if Bulgaria is not too behind in the 4th quarter.
- Considering the last appearances of Codi Miller-McIntyre for Saski Baskonia in the EuroLeague Basketball, the Spanish Liga ACB and Bulgaria, and also coach Barchovski's tasks to his players, the attacking of the "middles" markets by today's tune-ups is much less risky today.
- Others.

If both Vezenkov and Miller-McIntyre were absent today, some offered MoneyLine today's odds at that level would be honored by Kelly's Criterion.

Anyone preparing her/his participation in today's BG vs DE markets?

EuroBasket Qualification, BUL vs DE at the moment, Half-time.

A very fast practical drop of more than 14 points (17 in some cases in some agencies) of the "Points Spread" offered lines by the agencies within minutes - from high level since somewhere in the 1st and also included implosion start of the 2nd quarter. When it comes to the sporting things outside of the offered lines, the maximum spread was something like 15 pts so far.
A much larger drop in the "Point Spread" offered lines before the stability phase was initialized than the Model "predicted" before the match. And Bulgaria even overcame (-15) pts and even leads at the moment, but I think the hypothetical value on the MoneyLine market at Half-Time doesn't cover some risks.
That's why in my opinion the preservation of the market achievements in the first 2 quarters so far should be leading during the 3rd and 4th quarter.

What an encounter, indeed!

LeBron James made it to 40.000 points yesterday night. I was studying the stats of various players and came to the following conclusions:

- The best basketball player as we speak is of course Giannis Antetokounmpo. But I believe he will never make it to take out LeBron James from the top of the NBA scorers. Giannis is fragile while LeBron is anti fragile. Giannis style of play can not be maintained after a certain age, say 35-36. Unless he makes a huge turn in his career to become a point guard in the next 5-6 years. I m sure he already knows it and he is working on it.
- The player that has the stats to take out LeBron's record in the future is Luka Doncic. He is by far the biggest talent the NBA has seen. He is also anti fragile and he has the time to beat all records aged just 25. Unless something bad happens with his career, Luka Doncic will positevely surprise all basketball friends.
- Another great talent is of course Joel Embiid. He has the stats and the time to beat LeBron James, but he is fragile. The injuries he suffered makes it clear that unfortunately his career will not be uninterrupted.

It's so so curious the high number of corners, when the favourite team starts losing a match.
Now we can see in Manchester City vs Manchester United. But this happens always.
And it's so so curious how the number of corners stop to rise, when the favourite finally score.

Now it's 12-2 in corners in the intermediate.

Let's see if i'm right in the second half, or i have to eat my words

btw, i'm not sure, if bookmakers take this right or not.

arbusers wrote: ↑

Sun Mar 03, 2024 6:05 am

LeBron James made it to 40.000 points yesterday night. I was studying the stats of various players and came to the following conclusions:

- The best basketball player as we speak is of course Giannis Antetokounmpo. But I believe he will never make it to take out LeBron James from the top of the NBA scorers. Giannis is fragile while LeBron is anti fragile. Giannis style of play can not be maintained after a certain age, say 35-36. Unless he makes a huge turn in his career to become a point guard in the next 5-6 years. I m sure he already knows it and he is working on it.
- The player that has the stats to take out LeBron's record in the future is Luka Doncic. He is by far the biggest talent the NBA has seen. He is also anti fragile and he has the time to beat all records aged just 25. Unless something bad happens with his career, Luka Doncic will positevely surprise all basketball friends.
- Another great talent is of course Joel Embiid. He has the stats and the time to beat LeBron James, but he is fragile. The injuries he suffered makes it clear that unfortunately his career will not be uninterrupted.


Giannis Luka.png

Excellent material that hinted at different things for future modeling, thank you!

CharlieSheen99 wrote: ↑

Sun Mar 03, 2024 4:31 pm

It's so so curious the high number of corners, when the favourite team starts losing a match.
Now we can see in Manchester City vs Manchester United. But this happens always.
And it's so so curious how the number of corners stop to rise, when the favourite finally score.

Now it's 12-2 in corners in the intermediate.

Let's see if i'm right in the second half, or i have to eat my words

btw, i'm not sure, if bookmakers take this right or not.

The bookies (almost) always play their strict profitable game provided that their management is doing it correctly. The almost unlimited money behind them, and the almost unlimited long-run game possibilities give them a blow-minded advantage over regular individuals like us.
However, I assure you, that the data & stats processing and observing various things is causing the comeback of serious long-term-oriented individuals in the battle for a profitable long-term game.

You mentioned the corner markets, imagine if a player injected X money before the match in the "Y corners" market and opened statistically proven in the long-term period (by someone's Model), let's say Z positions when Carlo Ancelotti ordered Luka Modrić to enter the pitch, and also be open-minded for various Sigmas. See very interesting examples in Spain here: have-a-break-have-a-relaxing-chatting-b ... 494#p98494

Aussie Basketball, Play-Offs at the moment,
Considering the deuce 42:42 at half-time and the current Stats, in my opinion, there are some possibilities for attacks on the safely most probable "middles" positions in the next two quarters in the encounter Illawarra Hawks vs NZ Breakers, but preferably more safely in terms of staking positions, having in mind the next game of that Series. 31:17 and 11:25 in past quarters taught us additional warnings (due to 0.500 & 0.464 percentages in the regular season).

Strong defensive game tactics by both teams in the 4th quarter, zero scored points for two minutes and six for three ones. O/U markets possibilities now.

The following photo explains very well some of the phenomena we see in the US, and I am not talking about basketball only. I believe it is a matter of time until we see Luka Doncic and the Slovenia flag.

EuroLeague Basketball,
Maccabi Tel Aviv vs Žalgiris Kaunas, half-time at the moment,
I mean the temporary result of 33:33 moments ago and also the attempts to maneuver of coach Trinchieri were the best value moments so far, considering one of the proposed initial lines by the Asians of (-6.5) points.
The next market positions after half-time would be secondary related to this. Exactly in these kinds of events (Maccabi Tel Aviv's slow reaction involved), after all, Kelly will be Kelly even in the tradin' approach.

Now is the time for self-criticism. I firmly believe that the pre-match-calculated field {X;Y} errors are amongst our best teachers.

The idea behind my post is that unchanged criteria for a model's tune-ups for a very long time would cause some factors to be overlooked.
The same effect is possible after too early use of new things, not been tested for a long time.
If you are lucky enough to have your new model fail several times in a very short period of time, you will immediately know exactly what is wrong and what needs to be improved. If you're not that lucky, a suboptimal basis for your model can subtly eat up a partial portion of your income for a very long time before you know it.

Same case as Zeno's pseudo-paradox of Achilles and the Tortoise. If you somehow ignore the Time factor, you will decide, as Zeno of Elea did, that Achilles never reaches the Tortoise, when in fact the time it takes Achilles to reach the Tortoise is equal to the Tortoise's initial starting distance advantage over Achilles (in meters) divided by the speed difference (meters/second) between fast-running Achilles and the slow-moving animal.

Well, I was lucky enough to see some errors very quickly, as I said above. I changed some things after very short tests and as a consequence had to use stop-loss/hedge-operations on three occasions within only a few hours:
- Oklahoma @ Spurs 118:132, improperly weighted data and poor preliminary evaluations of some San Antonio players, and that day the Model focused mainly on Wembanyama and coach Popovich. Overestimating City Thunder's capabilities that day. The real point difference wasn't among the leading calculated ones in terms of probabilities;
- Real Madrid vs Panathinaikos 86:97, my initial plan was mainly targeting covering some specific zones of the most probable pre-calculated difference area {+8;-3}. The real (-11) is very far from that. The same errors as San Antonio above - one real pre-estimation for Lessort (Pretty OK) and poor for other players.
The pre-match value in visitors' odds was clear and the Ataman's behavior in the fourth quarter was as predicted, but the value in ML was gone. And so on;
- Partizan Belgrade vs Anadolu Efes 100:90, similar settings, calculated probabilities for {+3;-1} and too far from the real one. After days of preparation especially for that match, serious data processed and lots of videos watched, the custom-made graphics pointed me me the very high probability for ending the bad recent Partizan series that day vs Efes. However, some weight was assigned to the way Partizan lost recently vs Bayern 78:79, vs. Virtus 84:88 and Monaco 70:85 and also the way Anadolu Efes behaved away vs. Real Madrid recently 130:126 (4 Overtimes). The final (+10) was far from the top calculated probabilities, but the graphics hinted to me at the exceptional probability of a minimal-points difference win.
Other errors of the new settings: there were not included in the Model some of the recent Partizan matches in the Adriatic League and also the internal Korać Cup (National Cup) tournament. And many more things to be improved.

The decision - the old stable version of the Model was restored the next day, and the testing of the new will be continued for a much longer time than expected.
NBA, Spurs @ Kings in a few hours, and later that day EuroLeague, Panathinaikos vs Lyon-Vill and Olimpia Armani Milan vs Partizan Belgrade are pre-loaded in the old Model.

In short - it's much better if the evolution of your market methods is stable, even if being slightly outdated.